This is the multi-page printable view of this section. Click here to print.

Return to the regular view of this page.

Compute

Leveraging the country’s top reliability, we conveniently and elastically provide optimal computing resources tailored to each use case.

1 - Virtual Server

1.1 - Overview

Service Overview

Virtual Server is a virtual server optimized for cloud computing that lets you freely allocate and use infrastructure resources provided by the server—such as CPU and memory—without having to purchase them individually, and you can allocate as much as you need, when you need it. In a cloud environment, you can use resources with optimized performance according to the user’s computing purposes such as development, testing, and application execution.

Features

  • Easy and convenient computing environment setup: Through a web-based Console, users can easily use Self Service from Virtual Server provisioning to resource management and cost management. * When using a Virtual Server, if you need to change the capacity of major resources such as CPU or Memory, you can easily scale up or down without operator intervention.

  • Various Service Types Offered: Provides virtualized vCore/Memory resources based on predefined server types (1~128 vCore).

    • General Virtual Server: Provides commonly used computing specs (up to 16vCore, 256GB)
    • Large-capacity Virtual Server: Provided when resources larger than the standard Virtual Server spec are required.
  • Strong Security Implementation: Through the Security Group service, control inbound/outbound traffic communicating with the external Internet or other VPC(Virtual Private Cloud) to securely protect the server. * Additionally, real-time monitoring enables stable operation of computing resources.

Service architecture diagram

Diagram
Figure. Virtual Server diagram

Provided features

Virtual Server provides the following features.

  • Auto Provisioning and Management: Provides Virtual Server provisioning, resource management, and cost management functions through a web-based console. * If you need to change the capacity of major resources such as CPU or Memory while using a Virtual Server, you can modify it immediately using the server type edit function.
  • Standard Server Type and Image Provision: Provides virtualized vCore/Memory resources according to the standard server type, and supplies a standard OS Image.
  • Storage Connection: Provides additional attached storage besides the OS disk. * Block Storage, File Storage, Object Storage can be additionally connected and used.
  • Network Connection: You can connect the Virtual Server’s standard subnet/IP settings and Public NAT IP. * Provides a local subnet connection for inter-server communication. * This task can be edited on the detail page.
  • Security Group Application: Through the Security Group service, control inbound/outbound traffic communicating with external internet or other VPCs to securely protect the server.
  • Monitoring: You can view monitoring information for computing resources such as CPU, Memory, and Disk through the Cloud Monitoring service.
  • Backup and Recovery: You can back up and recover Virtual Server Image using the Backup service.
  • Cost Management: You can create, stop, or terminate servers as needed, and because billing is based on actual usage time, you can verify costs according to usage.
  • ServiceWatch Service Integration: You can monitor data using the ServiceWatch service.

Components

Virtual Server provides standard server types and standard OS images. Users can select and use it according to the desired service scale.

Image

You can create and manage images. The main features are as follows.

  • Image creation: You can create an Image from the configuration of the Virtual Server you are using, and you can create an Image by uploading your Image file to Object Storage.
  • Create Shared Image: You can create an Image with Visibility set to Private as a Shared Image that can be shared.
  • Share with another Account: You can share the Image with another Account.
  • For instructions on creating and using images, see the How-to guides > Image document.

Keypair

To ensure a more secure OS login, we strengthen security by providing a Key Pair instead of the ID/Password entry method. The main features are as follows.

  • Keypair creation: Creates a user credential for connecting to the Virtual Server.
  • Retrieve Public Key: You can retrieve the public key by loading a file or by manually entering the public key. Keypair 생성 및 활용 방법은 How-to guides > Keypair문서를 참고하세요.

Server Group {#server-group}

Server Group 설정을 통해 Virtual Server 및 Virtual Server 생성 시 추가한 Block Storage를 랙(Rack) 및 호스트에 근접 또는 분산 배치 가능합니다. 주요 기능은 다음과 같습니다.

Server Group 생성: 동일 Server Group에 소속된 Virtual Server를 Anti-Affinity(분산배치), Affinity(근접배치), Partition(Virtual Server와 Block Storage 분산배치)로 설정할 수 있습니다.

OS Image version provided

The OS images provided by Virtual Server are as follows.

OS Image versionEoS Date
Alma Linux 8.102029-05-31
Alma Linux 9.62025-11-17
Oracle Linux 8.102029-07-31
Oracle Linux 9.62025-11-25
RHEL 8.102029-05-31
RHEL 9.42026-04-30
RHEL 9.62027-05-31
Rocky Linux 8.102029-05-31
Rocky Linux 9.62025-11-30
Ubuntu 22.042027-06-30
Ubuntu 24.042029-06-30
Windows 20162027-01-12
Windows 20192029-01-09
Windows 20222031-10-14
Table. Virtual Server provided OS Image version
Reference
  • Linux operating systems such as Alma Linux and Rocky Linux provide only even Minor versions, except for the final release version of a Major version. * This is a policy to ensure the stability and consistency of the SCP system. We recommend checking the EOS (End of Support) and EOL (End of Life) dates for the operating system, and, if necessary, applying new or additional individual packages to maintain a stable environment.

Server type

The server types supported by Virtual Server are as follows. For detailed information about server types, see Virtual Server 서버 타입.

Standard s1v2m4
구분
예시상세 설명
서버 타입StandardProvided server type classifications
  • Standard: Configured with the commonly used standard specifications (vCPU, Memory)
  • High Capacity: Large‑capacity server specifications beyond Standard
Server specificationss1Provided server type classification and generation
  • s1: s means standard specification, and 1 indicates the generation provided by Samsung Cloud Platform v2
  • s2: s means standard specification, and 2 indicates the generation provided by Samsung Cloud Platform v2
  • h2: h means high-capacity server specification, and 2 indicates the generation provided by Samsung Cloud Platform v2
Server specificationsv2vCore count
  • v2: 2 virtual cores
Server specificationsm4Memory capacity
  • m4: 4GB Memory
Table. Virtual Server server type

Constraints

Reference
  • When creating a Virtual Server with Rocky Linux or Oracle Linux, additional configuration is required for time synchronization (NTP:Network Time Protocol). * If created with a different image, it is set automatically and no additional configuration is required.
    For more details, please refer to Linux NTP 설정하기.
  • If RHEL and Windows Server were created before August 2025, the RHEL Repository and WKMS (Windows Key Management Service) settings need to be modified.
    For detailed information, refer to RHEL Repo 및 WKMS 설정하기.

Preceding Service

This is a list of services that must be pre-configured before creating the service. Please refer to the guide provided for each service and prepare in advance.

Service CategoryserviceDetailed description
NetworkingVPCA service that provides an isolated virtual network in a cloud environment
NetworkingSecurity GroupVirtual firewall that controls server traffic
Table. Virtual Server Pre-service

1.1.1 - Server type

Virtual Server server type

Virtual Server provides server types according to the intended use. Server types are composed of various combinations such as CPU, Memory, and Network Bandwidth. When creating a Virtual Server, the host server used for the Virtual Server is determined by the selected server type. Please select the server type according to the specifications of the application you want to run on the Virtual Server.

The server types supported by Virtual Server are as follows.

Standard s1v2m4
Category
ExampleDetailed description
Server typeStandard제공되는 서버 타입 구분
  • Standard: 일반적으로 사용되는 표준 사양(vCPU, Memory)으로 구성
  • High Capacity: Standard 이상의 대용량 서버으로 사양
서버 사양s1제공되는 서버 타입 구분 및 세대를 의미
  • 영문자: 서버 타입 구분
    • s: 일반 사양
    • h: 대용량 서버 사양
  • 숫자: Samsung Cloud Platform v2에서 제공하는 세대
    • 1, 2, 3: 1세대, 2세대, 3세대
서버 사양v2vCore 개수
  • v2: 2개의 가상 코어
서버 사양m4메모리 용량
  • m4: 4 GB Memory
Table. Virtual Server server type format
Notice
For s3 and h3 server types, they are available only on Samsung Cloud Platform for Enterprise and Samsung Cloud Platform for Samsung. Samsung Cloud Platform Sovereign is scheduled to be offered after November 2026.

s1 server type

The s1 server type of Virtual Server is provided with standard specifications (vCPU, Memory) and is suitable for various applications.

  • Samsung Cloud Platform v2’s first generation: Intel 3rd‑generation (Ice Lake) Xeon Gold 6342 Processor up to 3.3 GHz
  • Supports up to 16 vCPUs and 256 GB of memory
  • Maximum networking speed of 12.5 Gbps
구분서버 타입vCPUMemoryNetwork Bandwidth
Standards1v1m21 vCore2 GBMaximum 10 Gbps
Standards1v2m42 vCore4 GBUp to 10 Gbps
Standards1v2m82 vCore8 GBUp to 10 Gbps
Standards1v2m162 vCore16 GBUp to 10 Gbps
Standards1v2m242 vCore24 GBUp to 10 Gbps
Standards1v2m322 vCore32 GBUp to 10 Gbps
Standards1v4m84 vCore8 GBUp to 10 Gbps
Standards1v4m164 vCore16 GBUp to 10 Gbps
Standards1v4m324 vCore32 GBUp to 10 Gbps
Standards1v4m484 vCore48 GB최대 10 Gbps
Standards1v4m644 vCore64 GB최대 10 Gbps
Standards1v6m126 vCore12 GBUp to 10 Gbps
Standards1v6m246 vCore24 GBUp to 10 Gbps
Standards1v6m486 vCore48 GBUp to 10 Gbps
Standards1v6m726 vCore72 GBUp to 10 Gbps
Standards1v6m966 vCore96 GBUp to 10 Gbps
Standards1v8m168 vCore16 GBUp to 10 Gbps
Standards1v8m328 vCore32 GBUp to 10 Gbps
Standards1v8m648 vCore64 GBUp to 10 Gbps
Standards1v8m968 vCore96 GBUp to 10 Gbps
Standards1v8m1288 vCore128 GBUp to 10 Gbps
Standards1v10m2010 vCore20 GBUp to 10 Gbps
Standards1v10m4010 vCore40 GB최대 10 Gbps
Standards1v10m8010 vCore80 GB최대 10 Gbps
Standards1v10m12010 vCore120 GBUp to 10 Gbps
Standards1v10m16010 vCore160 GBUp to 10 Gbps
Standards1v12m2412 vCore24 GBUp to 12.5 Gbps
Standards1v12m4812 vCore48 GBMaximum 12.5 Gbps
Standards1v12m9612 vCore96 GBUp to 12.5 Gbps
Standards1v12m14412 vCore144 GBMaximum 12.5 Gbps
Standards1v12m19212 vCore192 GBMaximum 12.5 Gbps
Standards1v14m2814 vCore28 GBUp to 12.5 Gbps
Standards1v14m5614 vCore56 GBUp to 12.5 Gbps
Standards1v14m11214 vCore112 GBMaximum 12.5 Gbps
Standards1v14m16814 vCore168 GBUp to 12.5 Gbps
Standards1v14m22414 vCore224 GB최대 12.5 Gbps
Standards1v16m3216 vCore32 GB최대 12.5 Gbps
Standards1v16m6416 vCore64 GBUp to 12.5 Gbps
Standards1v16m12816 vCore128 GBUp to 12.5 Gbps
Standards1v16m19216 vCore192 GBMaximum 12.5 Gbps
Standards1v16m25616 vCore256 GBUp to 12.5 Gbps
Table. Virtual Server server type specifications - s1 server type

s2 server type

The Virtual Server s2 server type is offered with standard specifications (vCPU, Memory) and is suitable for various applications.

  • Second generation of Samsung Cloud Platform v2: Intel 4th‑generation (Sapphire Rapids) Xeon Gold 6448H processor up to 3.2 GHz
  • Supports up to 16 vCPUs and 256 GB of memory
  • Maximum networking speed of 12.5 Gbps
CategoryServer typeCPU vCoreMemoryNetwork Bandwidth(Gbps)
Standards2v1m21 vCore2 GBUp to 10 Gbps
Standards2v2m42 vCore4 GBUp to 10 Gbps
Standards2v2m82 vCore8 GBUp to 10 Gbps
Standards2v2m162 vCore16 GBUp to 10 Gbps
Standards2v2m242 vCore24 GBUp to 10 Gbps
Standards2v2m322 vCore32 GBUp to 10 Gbps
Standards2v4m84 vCore8 GBUp to 10 Gbps
Standards2v4m164 vCore16 GBUp to 10 Gbps
Standards2v4m324 vCore32 GBUp to 10 Gbps
Standards2v4m484 vCore48 GBUp to 10 Gbps
Standards2v4m644 vCore64 GBUp to 10 Gbps
Standards2v6m126 vCore12 GBUp to 10 Gbps
Standards2v6m246 vCore24 GBUp to 10 Gbps
Standards2v6m486 vCore48 GBUp to 10 Gbps
Standards2v6m726 vCore72 GBMaximum 10 Gbps
Standards2v6m966 vCore96 GBUp to 10 Gbps
Standards2v8m168 vCore16 GBUp to 10 Gbps
Standards2v8m328 vCore32 GBUp to 10 Gbps
Standards2v8m648 vCore64 GBUp to 10 Gbps
Standards2v8m968 vCore96 GBUp to 10 Gbps
Standards2v8m1288 vCore128 GBUp to 10 Gbps
Standards2v10m2010 vCore20 GBUp to 10 Gbps
Standards2v10m4010 vCore40 GBUp to 10 Gbps
Standards2v10m8010 vCore80 GBUp to 10 Gbps
Standards2v10m12010 vCore120 GBUp to 10 Gbps
Standards2v10m16010 vCore160 GBUp to 10 Gbps
Standards2v12m2412 vCore24 GB최대 12.5 Gbps
Standards2v12m4812 vCore48 GBUp to 12.5 Gbps
Standards2v12m9612 vCore96 GBUp to 12.5 Gbps
Standards2v12m14412 vCore144 GBUp to 12.5 Gbps
Standards2v12m19212 vCore192 GBUp to 12.5 Gbps
Standards2v14m2814 vCore28 GBUp to 12.5 Gbps
Standards2v14m5614 vCore56 GBUp to 12.5 Gbps
Standards2v14m11214 vCore112 GBUp to 12.5 Gbps
Standards2v14m16814 vCore168 GBUp to 12.5 Gbps
Standards2v14m22414 vCore224 GBUp to 12.5 Gbps
Standards2v16m3216 vCore32 GBUp to 12.5 Gbps
Standards2v16m6416 vCore64 GBUp to 12.5 Gbps
Standards2v16m12816 vCore128 GBUp to 12.5 Gbps
Standards2v16m19216 vCore192 GBUp to 12.5 Gbps
Standards2v16m25616 vCore256 GBUp to 12.5 Gbps
Table. Virtual Server server type specifications – s2 server type

s3 server type

The Virtual Server s3 server type is offered with standard specifications (vCPU, Memory) and is suitable for various applications.

  • Samsung Cloud Platform v2’s 3rd generation: Intel 6th‑generation (Granite Rapids) Xeon 6737P Processor up to 4.0 GHz
  • Supports up to 16 vCPUs and 256 GB of memory
  • Maximum networking speed of 12.5 Gbps
CategoryServer typeCPU vCoreMemoryNetwork Bandwidth(Gbps)
Standards3v1m21 vCore2 GBUp to 10 Gbps
Standards3v2m42 vCore4 GBMaximum 10 Gbps
Standards3v2m82 vCore8 GBUp to 10 Gbps
Standards3v2m162 vCore16 GBUp to 10 Gbps
Standards3v2m242 vCore24 GBMaximum 10 Gbps
Standards3v2m322 vCore32 GBUp to 10 Gbps
Standards3v4m84 vCore8 GBUp to 10 Gbps
Standards3v4m164 vCore16 GBUp to 10 Gbps
Standards3v4m324 vCore32 GBUp to 10 Gbps
Standards3v4m484 vCore48 GBMaximum 10 Gbps
Standards3v4m644 vCore64 GBUp to 10 Gbps
Standards3v6m126 vCore12 GBMaximum 10 Gbps
Standards3v6m246 vCore24 GBUp to 10 Gbps
Standards3v6m486 vCore48 GBUp to 10 Gbps
Standards3v6m726 vCore72 GBUp to 10 Gbps
Standards3v6m966 vCore96 GBMaximum 10 Gbps
Standards3v8m168 vCore16 GBMaximum 10 Gbps
Standards3v8m328 vCore32 GBUp to 10 Gbps
Standards3v8m648 vCore64 GBMaximum 10 Gbps
Standards3v8m968 vCore96 GBUp to 10 Gbps
Standards3v8m1288 vCore128 GBUp to 10 Gbps
Standards3v10m2010 vCore20 GBUp to 10 Gbps
Standards3v10m4010 vCore40 GBMaximum 10 Gbps
Standards3v10m8010 vCore80 GBMaximum 10 Gbps
Standards3v10m12010 vCore120 GBUp to 10 Gbps
Standards3v10m16010 vCore160 GBMaximum 10 Gbps
Standards3v12m2412 vCore24 GBUp to 12.5 Gbps
Standards3v12m4812 vCore48 GBUp to 12.5 Gbps
Standards3v12m9612 vCore96 GBUp to 12.5 Gbps
Standards3v12m14412 vCore144 GBMaximum 12.5 Gbps
Standards3v12m19212 vCore192 GBUp to 12.5 Gbps
Standards3v14m2814 vCore28 GBUp to 12.5 Gbps
Standards3v14m5614 vCore56 GBUp to 12.5 Gbps
Standards3v14m11214 vCore112 GBUp to 12.5 Gbps
Standards3v14m16814 vCore168 GBUp to 12.5 Gbps
Standards3v14m22414 vCore224 GBUp to 12.5 Gbps
Standards3v16m3216 vCore32 GBUp to 12.5 Gbps
Standards3v16m6416 vCore64 GBUp to 12.5 Gbps
Standards3v16m12816 vCore128 GBUp to 12.5 Gbps
Standards3v16m19216 vCore192 GBUp to 12.5 Gbps
Standards3v16m25616 vCore256 GBUp to 12.5 Gbps
표. Virtual Server 서버 타입 사양 - s3 서버 타입

h2 Server Type

The h2 server type of Virtual Server is offered with high-capacity server specifications and is suitable for applications that require large-scale data processing.

  • Second generation of Samsung Cloud Platform v2: Intel 4th‑generation (Sapphire Rapids) Xeon Gold 6448H processor up to 3.2 GHz
  • Supports up to 128 vCPUs and 1,536 GB of memory
  • Maximum networking speed of 25 Gbps
CategoryServer typevCPUMemoryNetwork Bandwidth
High Capacityh2v24m4824 vCore48 GBMaximum 25 Gbps
High Capacityh2v24m9624 vCore96 GBMaximum 25 Gbps
High Capacityh2v24m19224 vCore192 GBMaximum 25 Gbps
High Capacityh2v24m28824 vCore288 GBMaximum 25 Gbps
High Capacityh2v32m6432 vCore64 GBMaximum 25 Gbps
High Capacityh2v32m12832 vCore128 GBMaximum 25 Gbps
High Capacityh2v32m25632 vCore256 GBMaximum 25 Gbps
High Capacityh2v32m38432 vCore384 GBMaximum 25 Gbps
High Capacityh2v48m9648 vCore96 GBMaximum 25 Gbps
High Capacityh2v48m19248 vCore192 GBMaximum 25 Gbps
High Capacityh2v48m38448 vCore384 GBMaximum 25 Gbps
High Capacityh2v48m57648 vCore576 GBMaximum 25 Gbps
High Capacityh2v64m12864 vCore128 GBMaximum 25 Gbps
High Capacityh2v64m25664 vCore256 GBMaximum 25 Gbps
High Capacityh2v64m51264 vCore512 GBMaximum 25 Gbps
High Capacityh2v64m76864 vCore768 GBMaximum 25 Gbps
High Capacityh2v72m14472 vCore144 GBMaximum 25 Gbps
High Capacityh2v72m28872 vCore288 GBMaximum 25 Gbps
High Capacityh2v72m57672 vCore576 GBMaximum 25 Gbps
High Capacityh2v72m86472 vCore864 GBMaximum 25 Gbps
High Capacityh2v96m19296 vCore192 GBMaximum 25 Gbps
High Capacityh2v96m38496 vCore384 GBMaximum 25 Gbps
High Capacityh2v96m76896 vCore768 GBMaximum 25 Gbps
High Capacityh2v96m115296 vCore1152 GBMaximum 25 Gbps
High Capacityh2v128m256128 vCore256 GBMaximum 25 Gbps
High Capacityh2v128m512128 vCore512 GBMaximum 25 Gbps
High Capacityh2v128m1024128 vCore1024 GBMaximum 25 Gbps
High Capacityh2v128m1536128 vCore1536 GBMaximum 25 Gbps
Table. Virtual Server server type specifications - h2 server type

h3 Server Type

The h3 server type of Virtual Server is offered with high-capacity specifications and is suitable for applications that require large-scale data processing.

  • Samsung Cloud Platform v2’s 3rd generation: Intel 6th‑generation (Granite Rapids) Xeon 6738P Processor up to 4.1 GHz
  • Supports up to 128 vCPUs and 1,536 GB of memory
  • Maximum networking speed of 25G bps
CategoryServer typevCPUMemoryNetwork Bandwidth
High Capacityh3v24m4824 vCore48 GBMaximum 25 Gbps
High Capacityh3v24m9624 vCore96 GBMaximum 25 Gbps
High Capacityh3v24m19224 vCore192 GBMaximum 25 Gbps
High Capacityh3v24m28824 vCore288 GBMaximum 25 Gbps
High Capacityh3v32m6432 vCore64 GBMaximum 25 Gbps
High Capacityh3v32m12832 vCore128 GBMaximum 25 Gbps
High Capacityh3v32m25632 vCore256 GBMaximum 25 Gbps
High Capacityh3v32m38432 vCore384 GBMaximum 25 Gbps
High Capacityh3v48m9648 vCore96 GBMaximum 25 Gbps
High Capacityh3v48m19248 vCore192 GBMaximum 25 Gbps
High Capacityh3v48m38448 vCore384 GBMaximum 25 Gbps
High Capacityh3v48m57648 vCore576 GBMaximum 25 Gbps
High Capacityh3v64m12864 vCore128 GBMaximum 25 Gbps
High Capacityh3v64m25664 vCore256 GBMaximum 25 Gbps
High Capacityh3v64m51264 vCore512 GBMaximum 25 Gbps
High Capacityh3v64m76864 vCore768 GBMaximum 25 Gbps
High Capacityh3v72m14472 vCore144 GBMaximum 25 Gbps
High Capacityh3v72m28872 vCore288 GBMaximum 25 Gbps
High Capacityh3v72m57672 vCore576 GBMaximum 25 Gbps
High Capacityh3v72m86472 vCore864 GBUp to 25 Gbps
High Capacityh3v96m19296 vCore192 GBMaximum 25 Gbps
High Capacityh3v96m38496 vCore384 GBMaximum 25 Gbps
High Capacityh3v96m76896 vCore768 GBMaximum 25 Gbps
High Capacityh3v96m115296 vCore1152 GBMaximum 25 Gbps
High Capacityh3v128m256128 vCore256 GBMaximum 25 Gbps
High Capacityh3v128m512128 vCore512 GBMaximum 25 Gbps
High Capacityh3v128m1024128 vCore1024 GBMaximum 25 Gbps
High Capacityh3v128m1536128 vCore1536 GBMaximum 25 Gbps
Table. Virtual Server server type specifications - h3 server type

1.1.2 - Monitoring Metrics

Cloud Monitoring service termination notice

According to Samsung Cloud Platform’s policy, the Cloud Monitoring service is scheduled to be discontinued in September 2026.
Starting after the September 2026 release, resource monitoring of the Samsung Cloud Platform via Cloud Monitoring will no longer be possible.

With the new alternative service, you can continuously perform resource monitoring by leveraging ServiceWatch released in October 2025.
ServiceWatch provides more modern and powerful features, replacing Cloud Monitoring to deliver a seamless monitoring environment.

Detailed information about ServiceWatch can be found in the ServiceWatch Overview.

Virtual Server Monitoring Metrics

The table below shows the monitoring metrics of Virtual Server that can be viewed through Cloud Monitoring. For detailed usage of Cloud Monitoring, refer to the Cloud Monitoring guide.

Provides basic monitoring metrics even without installing an agent, as shown below table. Please check the Virtual Server monitoring metrics (default). Additionally, the metrics that can be viewed by installing the Agent are in the table below. Virtual Server additional monitoring metrics (Agent installation required)** Please refer to it.

For Windows OS, memory-related metrics can only be viewed after installing the Agent.

Performance itemsDetailed descriptionunit
Memory Total [Basic]bytes of usable memorybytes
Memory Used [Basic]Current memory usage in bytesbytes
Memory Swap In [Basic]bytes of the replaced memorybytes
Memory Swap Out [Basic]bytes of the replaced memorybytes
Memory Free [Basic]bytes of unused memorybytes
Disk Read Bytes [Basic]Read bytesbytes
Disk Read Requests [Basic]Number of read requestscnt
Disk Write Bytes [Basic]write bytesbytes
Disk Write Requests [Basic]Number of write requestscnt
CPU Usage [Basic]Average system CPU usage over 1 minute%
Instance State [Basic]Instance statusstate
Network In Bytes [Basic]Received bytesbytes
Network In Dropped [Basic]Incoming packet dropcnt
Network In Packets [Basic]Number of received packetscnt
Network Out Bytes [Basic]sent bytesbytes
Network Out Dropped [Basic]Transmit packet dropcnt
Network Out Packets [Basic]Number of transmitted packetscnt
Table. Virtual Server Monitoring Metrics (Provided by default)
Performance itemsDetailed descriptionunit
Core Usage [IO Wait]Ratio of CPU time spent in wait state (disk wait)%
Core Usage [System]Proportion of CPU time spent in kernel space%
Core Usage [User]Proportion of CPU time spent in user space%
CPU CoresNumber of CPU cores on the hostcnt
CPU Usage [Active]Percentage of CPU time used other than Idle and IOWait states%
CPU Usage [Idle]It is the proportion of CPU time spent in idle state.%
CPU Usage [IO Wait]This is the proportion of CPU time spent in a waiting state (disk wait).%
CPU Usage [System]Percentage of CPU time used by the kernel%
CPU Usage [User]Percentage of CPU time used in user space%
CPU Usage/Core [Active]Percentage of CPU time used other than Idle and IOWait states%
CPU Usage/Core [Idle]It is the proportion of CPU time spent in idle state.%
CPU Usage/Core [IO Wait]This is the proportion of CPU time spent in a waiting state (disk wait).%
CPU Usage/Core [System]Percentage of CPU time used by the kernel%
CPU Usage/Core [User]Percentage of CPU time used in user space%
DiskCPU Usage [IO Request]Proportion of CPU time during which I/O requests to the device were executed%
Disk Queue Size [Avg]The average queue length of requests executed for the device.num
Disk Read BytesThe number of bytes read per second from the device.bytes
Disk Read Bytes [Delta Avg]Average of system.diskio.read.bytes_delta for individual disksbytes
Disk Read Bytes [Delta Max]Maximum system.diskio.read.bytes_delta of individual disksbytes
Disk Read Bytes [Delta Min]Minimum of system.diskio.read.bytes_delta for individual disksbytes
Disk Read Bytes [Delta Sum]Sum of system.diskio.read.bytes_delta of individual disksbytes
Disk Read Bytes [Delta]Delta of the system.diskio.read.bytes value for each Diskbytes
Disk Read Bytes [Success]Total bytes successfully readbytes
Disk Read RequestsNumber of read requests to the disk device per secondcnt
Disk Read Requests [Delta Avg]Average of system.diskio.read.count_delta for individual diskscnt
Disk Read Requests [Delta Max]Maximum system.diskio.read.count_delta for individual diskscnt
Disk Read Requests [Delta Min]Minimum of system.diskio.read.count_delta for individual diskscnt
Disk Read Requests [Delta Sum]Sum of the system.diskio.read.count_delta of individual diskscnt
Disk Read Requests [Success Delta]Delta of system.diskio.read.count for each Diskcnt
Disk Read Requests [Success]Total number of successful readscnt
Disk Request Size [Avg]It is the average size of requests executed on the device (unit: sectors).num
Disk Service Time [Avg]Average service time (milliseconds) of input requests executed on the device.ms
Disk Wait Time [Avg]Average time taken for requests executed on the supported device.ms
Disk Wait Time [Read]Average disk wait timems
Disk Wait Time [Write]Average disk wait timems
Disk Write Bytes [Delta Avg]Average of system.diskio.write.bytes_delta for each diskbytes
Disk Write Bytes [Delta Max]Maximum system.diskio.write.bytes_delta of individual disksbytes
Disk Write Bytes [Delta Min]Minimum of system.diskio.write.bytes_delta for individual disksbytes
Disk Write Bytes [Delta Sum]Sum of system.diskio.write.bytes_delta for individual disksbytes
Disk Write Bytes [Delta]Delta of the system.diskio.write.bytes value for each Diskbytes
Disk Write Bytes [Success]Total number of bytes successfully writtenbytes
Disk Write RequestsNumber of write requests to the disk device per secondcnt
Disk Write Requests [Delta Avg]Average of system.diskio.write.count_delta for individual diskscnt
Disk Write Requests [Delta Max]Maximum system.diskio.write.count_delta of individual diskscnt
Disk Write Requests [Delta Min]Minimum of system.diskio.write.count_delta for individual diskscnt
Disk Write Requests [Delta Sum]Sum of system.diskio.write.count_delta for individual diskscnt
Disk Write Requests [Success Delta]Delta of system.diskio.write.count for each Diskcnt
Disk Write Requests [Success]Total number of successful writescnt
Disk Writes BytesBytes per second written to the devicebytes
Filesystem Hang Checkfilesystem(local/NFS) hang check(normal:1, abnormal:0)status
Filesystem NodesTotal number of file nodes in the file system.cnt
Filesystem Nodes [Free]It is the total number of available file nodes in the file system.cnt
Filesystem Size [Available]Disk space (bytes) available to unauthorized usersbytes
Filesystem Size [Free]Available disk space (bytes)bytes
Filesystem Size [Total]Total disk space (bytes)bytes
Filesystem UsageUsed disk space percentage%
Filesystem Usage [Avg]Average of individual filesystem.used.pct%
Filesystem Usage [Inode]iNode usage rate%
Filesystem Usage [Max]Maximum among individual filesystem.used.pct%
Filesystem Usage [Min]Min among individual filesystem.used.pct%
Filesystem Usage [Total]-%
Filesystem UsedUsed disk space (bytes)bytes
Filesystem Used [Inode]iNode usagebytes
Memory FreeTotal available memory (bytes)bytes
Memory Free [Actual]Actual usable Memory (bytes)bytes
Memory Free [Swap]Available Swap memorybytes
Memory TotalTotal Memorybytes
Memory Total [Swap]Total Swap memory.bytes
Memory UsagePercentage of used memory%
Memory Usage [Actual]Percentage of memory actually used%
Memory Usage [Cache Swap]cached swap usage%
Memory Usage [Swap]Percentage of used Swap memory%
Memory UsedUsed Memorybytes
Memory Used [Actual]Actual memory used (bytes)bytes
Memory Used [Swap]Used Swap memorybytes
CollisionsNetwork collisioncnt
Network In BytesNumber of received bytesbytes
Network In Bytes [Delta Avg]Average of system.network.in.bytes_delta for individual networksbytes
Network In Bytes [Delta Max]Maximum system.network.in.bytes_delta for each Networkbytes
Network In Bytes [Delta Min]Minimum of system.network.in.bytes_delta for each networkbytes
Network In Bytes [Delta Sum]Sum of system.network.in.bytes_delta for individual networksbytes
Network In Bytes [Delta]Delta of received byte countbytes
Network In DroppedNumber of deleted packets among incoming packetscnt
Network In ErrorsNumber of errors during receptioncnt
Network In PacketsNumber of received packetscnt
Network In Packets [Delta Avg]Average of system.network.in.packets_delta for individual Networkscnt
Network In Packets [Delta Max]Maximum of system.network.in.packets_delta for each Networkcnt
Network In Packets [Delta Min]Minimum of system.network.in.packets_delta for each Networkcnt
Network In Packets [Delta Sum]Sum of system.network.in.packets_delta for individual Networkscnt
Network In Packets [Delta]Delta of received packet countcnt
Network Out BytesNumber of transmitted bytesbytes
Network Out Bytes [Delta Avg]Average of system.network.out.bytes_delta for individual networksbytes
Network Out Bytes [Delta Max]Maximum system.network.out.bytes_delta for each Networkbytes
Network Out Bytes [Delta Min]Minimum of system.network.out.bytes_delta for individual Networksbytes
Network Out Bytes [Delta Sum]Sum of system.network.out.bytes_delta for individual Networksbytes
Network Out Bytes [Delta]Delta of transmitted byte countbytes
Network Out DroppedNumber of deleted packets among outgoing packetscnt
Network Out ErrorsNumber of errors during transmissioncnt
Network Out PacketsNumber of transmitted packetscnt
Network Out Packets [Delta Avg]Average of system.network.out.packets_delta for each Networkcnt
Network Out Packets [Delta Max]Maximum system.network.out.packets_delta for each Networkcnt
Network Out Packets [Delta Min]Minimum of system.network.out.packets_delta for each Networkcnt
Network Out Packets [Delta Sum]Sum of system.network.out.packets_delta for individual networkscnt
Network Out Packets [Delta]Delta of transmitted packet countcnt
Open Connections [TCP]All open TCP connectionscnt
Open Connections [UDP]All open UDP connectionscnt
Port UsageConnectable port utilization%
SYN Sent SocketsNumber of sockets in SYN_SENT state (when connecting from local to remote)cnt
Kernel PID Maxkernel.pid_max valuecnt
Kernel Thread Maxkernel.threads-max valuecnt
Process CPU UsagePercentage of CPU time consumed by the process since the last update%
Process CPU Usage/CorePercentage of CPU time used by the process since the last event%
Process Memory UsageProportion of main memory (RAM) occupied by a process%
Process Memory UsedResident Set size. The amount of memory a process occupies in RAM.bytes
Process PIDprocess pidpid
Process PPIDParent process PIDpid
Processes [Dead]Number of deadProcessescnt
Processes [Idle]idle Processes countcnt
Processes [Running]running Processes countcnt
Processes [Sleeping]sleeping processes countcnt
Processes [Stopped]stopped processes countcnt
Processes [Total]Total number of processescnt
Processes [Unknown]Number of processes whose status cannot be retrieved or is unknowncnt
Processes [Zombie]Number of zombie processescnt
Running Process Usageprocess usage%
Running ProcessesNumber of running processescnt
Running Thread UsageThread usage rate%
Running ThreadsTotal number of threads running in running processescnt
Context Switchescontext switch count (per second)cnt
Load/Core [1 min]The load over the last 1 minute divided by the number of corescnt
Load/Core [15 min]The load over the last 15 minutes divided by the number of corescnt
Load/Core [5 min]The load over the last 5 minutes divided by the number of corescnt
Multipaths [Active]External storage connection path status = active countcnt
Multipaths [Failed]External storage connection path status = failed countcnt
Multipaths [Faulty]External storage connection path status = faulty countcnt
NTP Offset lastsample’s measured offset (time difference between NTP server and local environment)num
Run Queue LengthExecution queue lengthnum
UptimeOS uptime (milliseconds)ms
Context Switchies CPUcontext switch count (per second)cnt
Disk Read Bytes [Sec]Number of bytes read in one second from a Windows logical disk
  • Windows only
cnt
Disk Read Time [Avg]Data read average time (seconds)
  • Windows only
sec
Disk Transfer Time [Avg]Disk average wait time (seconds)
  • Windows only
sec
Disk Write Bytes [Sec]Number of bytes written in one second on a Windows logical disk
  • Windows only
cnt
Disk Write Time [Avg]Average data write time (seconds)
  • Windows only
sec
Pagingfile Usagepaging file usage
  • Windows only
%
Pool Used [Non Paged]Nonpaged Pool usage in kernel memory
  • Windows only
bytes
Pool Used [Paged]Paged Pool usage in kernel memory
  • Windows only
bytes
Process [Running]Number of currently running processes
  • Windows only
cnt
Threads [Running]Number of currently running threads
  • Windows only
cnt
Threads [Waiting]Number of threads waiting for processor time
  • Windows only
cnt
Table. Additional monitoring metrics for Virtual Server (Agent installation required)

1.1.3 - ServiceWatch Metrics

Virtual Server sends metrics to ServiceWatch. The metrics provided by default monitoring are data collected at 5‑minute intervals. If detailed monitoring is enabled, you can view data collected at 1‑minute intervals.

Reference
For how to view metrics in ServiceWatch, refer to the ServiceWatch guide.

See How-to guides > ServiceWatch Enable Detailed Monitoring for instructions on enabling detailed monitoring of Virtual Server.

Basic Metrics

The following are the basic metrics for the Virtual Server namespace.

The indicators whose names are displayed in bold below are the indicators selected as key metrics among the default metrics provided by Virtual Server. The key metrics are used to build service dashboards that are automatically generated for each service in ServiceWatch. They can also be viewed on the Monitoring tab of the Virtual Server detail page.

Each metric provides guidance in the user guide on which statistical value is meaningful to query, and among the meaningful statistics, the values shown in bold text are the primary statistics. In the service dashboard or monitoring tab, you can view primary metrics using these primary statistics.

Performance item (indicator name)Detailed descriptionunitmeaningful statistics
Instance StateInstance status display
  • 1 - Active
  • 0 - Off
None
  • Total
CPU UsageCPU usagePercent
  • Average
  • Maximum
  • Minimum
Disk Read BytesBytes read from block device (bytes)Bytes
  • Total
  • Average
  • Maximum
  • Minimum
Disk Read RequestsNumber of read requests on a block deviceCount
  • Total
  • Average
  • Maximum
  • Minimum
Disk Write BytesWrite capacity (bytes) on block deviceBytes
  • Total
  • Average
  • Maximum
  • Minimum
Disk Write RequestsNumber of write requests on block deviceCount
  • Total
  • Average
  • Maximum
  • Minimum
Network In BytesReceived bytes on the network interfaceBytes
  • Total
  • Average
  • Maximum
  • Minimum
Network In DroppedNumber of packet drops received on the network interfaceCount
  • Total
  • Average
  • Maximum
  • Minimum
Network In PacketsNumber of packets received on the network interfaceCount
  • Total
  • Average
  • Maximum
  • Minimum
Network Out BytesData transmitted on the network interface (bytes)Bytes
  • Total
  • Average
  • Maximum
  • Minimum
Network Out DroppedNumber of packet drops transmitted from the network interfaceCount
  • Total
  • Average
  • Maximum
  • Minimum
Network Out PacketsNumber of packets transmitted on the network interfaceCount
  • Total
  • Average
  • Maximum
  • Minimum
Table. Virtual Server basic metrics
Reference
Refer to the ServiceWatch Agent guide for how to collect metrics using the ServiceWatch Agent.

1.2 - How-to guides

Users can create the service by entering the required Virtual Server information and selecting detailed options through the Samsung Cloud Platform Console.

Create Virtual Server

You can create and use the Virtual Server service in the Samsung Cloud Platform Console.

To create a Virtual Server, follow these steps.

  1. All Services > Compute > Virtual Server menu, click it. 1. Navigate to the Service Home page of the Virtual Server.

  2. On the Service Home page, click the Create Virtual Server button. 2. Create Virtual Server page.

  3. On the Virtual Server Creation page, enter the information required to create the service and select detailed options.

    • Select the required information in the Image and version selection area.
      Category
      Required
      Detailed description
      ImageEssentialSelect the type of Image provided
      • Standard: Samsung Cloud Platform standard Image
        • Alma Linux, Oracle Linux, RHEL, Rocky Linux, Ubuntu, Windows
      • Custom: User-created Image
      • Kubernetes: Kubernetes Image
        • RHEL, Ubuntu
      • Marketplace: Image subscribed from Marketplace
      Image versionEssentialSelect version of the chosen Image
      • Provide version list of the provided server Image
      Table. Virtual Server Image and version selection input items
    • Enter or select the required information in the Service Information Input area.
      Category
      Required
      Detailed description
      Number of serversEssentialNumber of servers to create simultaneously
      • Only numeric input is allowed, enter a value between 1~100
      Service Type > Server TypeEssentialVirtual Server 서버 타입
      • Standard: standard specifications commonly used
      • High Capacity: large-capacity server specifications beyond Standard
      Service Type > Planned ComputeEssentialStatus of resources with Planned Compute set
      • In Use: Number of resources with Planned Compute that are currently in use
      • Configured: Number of resources with Planned Compute set
      • Coverage Preview: Amount applied per resource by Planned Compute
      Block StorageEssentialBlock Storage settings used by the server according to purpose
      • Basic OS: The area where the OS is installed and used
        • Enter the size in Units; the minimum size varies depending on the OS Image type
          • Alma Linux: Enter a value between 2 and 1,536
          • Oracle Linux: Enter a value between 7 and 1,536
          • RHEL: Enter a value between 2 and 1,536
          • Rocky Linux: Enter a value between 2 and 1,536
          • Ubuntu: Enter a value between 2 and 1,536
          • Windows: Enter a value between 4 and 1,536
        • SSD: High‑performance general volume
        • HDD: General volume
        • SSD/HDD_KMS: Additional encrypted volume using Samsung Cloud Platform KMS (Key Management Service) encryption keys
          • Encryption can only be applied at initial creation and cannot be changed afterward
          • Using the SSD_KMS disk type may cause performance degradation
        • SSD_Provisioned: SSD volume with configurable IOPS and Throughput
      • Additional: Use when additional user space beyond the OS area is needed
        • After selecting Use, enter the storage type and size
        • Click the + button to add storage, or the x button to delete (up to 25 can be added)
        • Enter the size in Units, with a value between 1 and 1,536
          • Since 1 Unit equals 8 GB, this creates 8 – 12,288 GB
        • SSD: High‑performance general volume
        • HDD: General volume
        • SSD/HDD_KMS: Additional encrypted volume using Samsung Cloud Platform KMS (Key Management Service) encryption keys
          • Encryption can only be applied at initial creation and cannot be changed afterward
          • Using the SSD_KMS disk type may result in performance degradation
        • SSD/HDD_MultiAttach: Volume that can be attached to two or more servers
        • SSD_Provisioned: SSD volume with configurable IOPS and Throughput
      • Delete on termination: When Delete on Termination is selected, the volume is terminated together with the server
        • Volumes with snapshots are not deleted even when Delete on termination is enabled
        • A multi‑attach volume can be deleted only when the server being removed is the last remaining server attached to the volume
      • Max IOPS: Enter a maximum IOPS value between 5,000 and 20,000
        • Cannot be set for disk types HDD, HDD_KMS, or HDD_MultiAttach
      • Max Throughput: Enter a maximum Throughput value between 250 and 1,000
        • Cannot be set for disk types HDD, HDD_KMS, or HDD_MultiAttach
      Server GroupSelectionAfter selecting Use to set Anti-Affinity (distributed placement), Affinity (proximate placement), Partition (distributed placement of Virtual Server and Block Storage) for servers belonging to the same Server Group
      • , select the Server Group
      • , choose Create New to create a Server Group
      Place servers belonging to the same Server Group according to the selected policy using a Best Effort method
      The policy is selected from Anti-Affinity (distributed placement), Affinity (proximate placement), or Partition (distributed placement of Virtual Server and Block Storage)
      Table. Virtual Server service information input items
      Caution
      When using the Partition (distributed placement of Virtual Server and Block Storage) policy among Server Group policies, you cannot allocate additional Block Storage Volumes after creating a Virtual Server, so create all required Block Storage during the Virtual Server creation stage.
    • Required Information Input area, please enter or select the necessary information.
      Category
      Required
      Detailed description
      Server nameEssentialEnter a name to distinguish the server when the selected number of servers is 1
      • Set the hostname to the entered server name
      • Enter using English letters, numbers, spaces, and special characters (-, _) within 63 characters
      Network Settings > Create New Network PortEssentialSet the network where the Virtual Server will be installed
      • VPC name: Select a pre‑created VPC
      • General Subnet: Select a pre‑created general Subnet
        • IP can be auto‑generated or manually entered, and if Enter is selected, the user can input the IP directly
        • NAT: Available only when there is a single server and the VPC is attached to an Internet Gateway. Checking Use allows selection of a NAT IP
        • NAT IP: Select a NAT IP
          • If no NAT IP is available to select, click the Create New button to generate a Public IP
          • Refresh button: Click to view and select the created Public IP
          • Creating a Public IP incurs charges according to the Public IP pricing policy
      • Local Subnet (optional): Select Use for the local Subnet
        • It is not a required element for creating the service
        • A pre‑created local Subnet must be selected
        • IP can be auto‑generated or manually entered, and selecting Enter allows the user to input the IP directly
      Network Settings > Specify Existing Network PortEssentialSet the network where the Virtual Server will be installed
      • VPC: Select a pre-created VPC
      • General Subnet: Select a pre-created General Subnet and Port
        • NAT: Available only when there is a single server and the VPC has an Internet Gateway attached. When selected, you can choose a NAT IP
        • NAT IP: Select the NAT IP
          • If there is no NAT IP to select, click the Create New button to generate a Public IP
          • Refresh button to view and select the created Public IP
      • Local Subnet (optional): Select Use for the local subnet
        • Select a pre-created local Subnet and Port
      Security GroupSelectionSettings required to connect to the server
      • Select: Up to 5 pre‑created Security Groups can be selected
      • Create New: If there is no applicable Security Group, create one separately in the Security Group service
      • If a Security Group is not set, all connections are blocked, so you must configure it to allow the necessary connections
      KeypairEssentialUser authentication method to use when connecting to the server
      • Default login account list by OS
        • Alma Linux: almalinux
        • Oracle Linux: cloud-user
        • RHEL: cloud-user
        • Rocky Linux: rocky
        • Ubuntu: ubuntu
        • Windows: sysadmin
      Table. Virtual Server required information input fields
      Reference
      • The default OS account can serve as the default user account in a cloud environment and can be used for initial configuration via cloud-init and SCP-related functions.
      • If you delete or change the primary account, related Samsung Cloud Platform functions may not work properly, so deletion or modification is not recommended.
    • Additional Information Input area, enter or select the required information.
      Category
      Required
      Detailed description
      LockSelectionLock usage setting
      • When you use Lock, it prevents actions such as terminating, starting, or stopping the server, thereby avoiding malfunctions caused by mistakes
      Init scriptSelectionScript executed when the server starts
      • The init script must be written as a Batch script for Windows, a Shell script for Linux, or cloud‑init, depending on the image type.
      • Up to 45,000 bytes can be entered
      tagSelectionAdd Tag
      • Up to 50 can be added per resource
      • After clicking the Add Tag button, input or select Key, Value values
      Table. Virtual Server additional information input fields
  4. Summary Verify the detailed information and estimated charges generated in the panel, then click the Create button.

  5. When the popup indicating creation opens, click the Confirm button.

    • When creation is complete, verify the created resources on the Virtual Server List page.
information
  • When entering the server name, if spaces and special characters (_) are used, the OS hostname will have those spaces and special characters (_) converted to the special character (-) and set. * Refer to this when setting the OS hostname.
    • Example: If the server name is ‘server name_01’, the OS hostname is set to ‘server-name-01’.
  • If you need to manage server names uniquely, specify a different server name (Prefix) when creating them.
    • When creating a server, because the numbering does not automatically increment based on the server name (Prefix), a Virtual Server with the same name can be created.
    • Example: If you first create two Virtual Servers using the server name (Prefix) ’test’, ’test-1’ and ’test-2’ will be created. * After that, even if you create two Virtual Servers using the prefix ’test’ again, ’test-1’ and ’test-2’ will be created.
Reference
  • When creating a Virtual Server with Rocky Linux or Oracle Linux, additional configuration is required for time synchronization (NTP:Network Time Protocol). * For detailed information, refer to Linux NTP 설정하기.
  • If RHEL and Windows Server were created before July 2025, the RHEL Repository and WKMS (Windows Key Management Service) settings need to be modified. * For more information, refer to RHEL Repo 및 WKMS 설정하기.

View Virtual Server Details

The Virtual Server service lets you view and modify the complete resource list and detailed information. The Virtual Server Details page consists of Details, Monitoring, Tags, Activity History tabs.

To view detailed information about the Virtual Server service, follow these steps.

  1. All Services > Compute > Virtual Server Click the menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. On the Service Home page, click the Virtual Server menu. 2. Go to the Virtual Server List page.
  3. On the Virtual Server List page, click the resource to view detailed information. 3. Navigate to the Virtual Server Details page.
    • Virtual Server Details page displays status information and additional feature information, and consists of Details, Monitoring, Tags, Activity Log tabs.
    • For detailed information about Virtual Server Additional Features, please refer to Virtual Server 관리 부가 기능.
      CategoryDetailed description
      Virtual Server statusState of a user-created Virtual Server
      • Build: State where a Build command has been issued
      • Building: Build in progress
      • Networking: Server creation process in progress
      • Scheduling: Server creation process in progress
      • Block_Device_Mapping: Connecting Block Storage during server creation
      • Spawning: State where the server creation process is ongoing
      • Active: Available state
      • Powering_off: State when a stop request is made
      • Deleting: Server deletion in progress
      • Reboot_Started: Reboot in progress state
      • Error: Error state
      • Migrating: State where the server is migrating to another host
      • Reboot: State where a Reboot command has been issued
      • Rebooting: Reboot in progress
      • Rebuild: State where a Rebuild command has been issued
      • Rebuilding: State when a Rebuild request is made
      • Rebuild_Spawning: State where the Rebuild process is ongoing
      • Resize: State where a Resize command has been issued
      • Resizing: Resize in progress
      • Resize_Prep: State when a server type modification is requested
      • Resize_Migrating: State where the server is moving to another host while resizing
      • Resize_Migrated: State where the server has completed moving to another host during resize
      • Resize_Finish: Resize completed
      • Revert_Resize: Resize or migration of the server failed for some reason. The target server is cleaned up and the original source server is restarted
      • Shutoff: State when powering off is completed
      • Verity_ Resize: State after Resize_Prep is performed following a server type modification request, where the server type is finalized or can be reverted
      • Resize_Reverting: State when a server type revert request is made
      • Resize_Confirming: State where the server’s Resize request is being confirmed
      Server controlButton to change server status
      • Start: Start a stopped server
      • Stop: Stop a running server
      • Restart: Restart a running server
      Image generationCreate a user Image from the current server’s Image
      Console logCurrent server console log view
      Create dumpCreate a dump of the current server
      • The dump file is created inside the Virtual Server
      RebuildThe OS area data and settings of the existing Virtual Server are deleted, and it is rebuilt as a new server
      Service cancellationCancel service button
      Table. Virtual Server status information and additional features

Detailed Information

On the Virtual Server List page, you can view detailed information of the selected resource and, if needed, modify the information.

CategoryDetailed description
serviceService Name
Resource TypeResource Type
SRNUnique resource ID in Samsung Cloud Platform
  • In Virtual Server, it refers to the Virtual Server SRN
Resource nameresource name
  • In the Virtual Server service it refers to the Virtual Server name
Resource IDUnique resource ID in the service
ConstructorUser who created the service
Creation Date/TimeService creation date and time
EditorUser who edited the service information
Modification timestampDate and time the service information was modified
Server nameServer name
  • Edit Click the button to change the name
  • When changing the server name, the OS hostname is not changed; only the information within the Samsung Cloud Platform Console is updated
  • Modification is not allowed for Virtual Servers created from other resources
Server typevCPU, memory information display
  • If you need to change to a different server type, click the Edit button to configure
Image nameServer OS Image and version
  • Image version and Image selectable by build date
LockIndicates whether Lock is enabled or disabled
  • If you need to change the Lock property value, click the Edit button to set it
Server groupServer group name to which the server belongs
  • If the server group is not used, it will not be displayed.
Keypair nameServer authentication information set by the user
  • The default login accounts for each OS are as follows.
    • Alma Linux: almalinux
    • Oracle Linux: cloud-user
    • RHEL: cloud-user
    • Rocky Linux: rocky
    • Ubuntu: ubuntu
    • Windows: sysadmin
Planned ComputeResource status with Planned Compute configured
ServiceWatch detailed monitoringDisplay whether ServiceWatch detailed monitoring is enabled
  • To enable ServiceWatch detailed monitoring, click the Edit button to configure
  • Not provided for Auto-Scaling Groups and Virtual Servers created in the Marketplace
networkNetwork information of the Virtual Server
  • VPC, standard Subnet, IP and status, Public NAT IP and status, Private NAT IP and status, Virtual IP, Security Group
  • If an IP change is required, click the Edit button to configure
    • Editing is possible only when the Virtual Server status is other than Active, Shutoff
    • For the default port, Default is displayed next to the IP, and it cannot be detached
  • If a Security Group change is required, click the Edit button to configure
  • If a Virtual IP change is required, it can be edited on the Virtual IP Management tab of the Networking > VPC > Subnet Details page
  • Add as a new network port: select a standard Subnet and IP
    • You can select a different standard Subnet within the same VPC
    • Enter the IP manually when editing
  • Add using an existing network port: select a pre‑created standard Subnet and port
Local SubnetLocal Subnet information of the Virtual Server
  • Local Subnet, Local Subnet IP, Security Group name, Virtual IP
  • If a Security Group change is required, you can click the Edit button to configure it
  • Add as a new network port: select the local Subnet and IP
    • After selecting another regular Subnet within the same VPC, enter the IP
  • Add to an existing network port: select a pre‑created local Subnet and port
Block StorageBlock Storage information attached to the server
  • Volume ID, Volume Name, Disk Type, Capacity, Connection Information, Type, Delete on termination, Status
  • Add: Connect additional Block Storage when needed
  • Edit Delete on termination: Modify the Delete on termination value of the selected Block Storage in the list
  • More > Detach: Detach the connection of the selected Block Storage from the list
    • Detachment is not possible for OS default Storage
Table. Virtual Server detailed information tab items

Monitoring

On the Virtual Server List page, you can monitor the ServiceWatch metrics of the selected resources. In the Monitoring tab, you can view monitoring charts for the Virtual Server, and each chart is based on the available Service Watch metrics.

CategoryDetailed description
Period setting areaPeriod selection applied to the chart
  • Indicator query can be set from now up to a maximum of 455 days
Time Zone Settings AreaSelect the time zone applied to the chart
Reset buttonReset all modifications and settings in the chart.
Refresh Settings AreaSelect chart refresh interval
  • Refresh button displays information anew based on the current time
  • Click the refresh interval to select the desired interval: Off, 10 seconds, 1 minute, 2 minutes, 5 minutes, 15 minutes
Go to the service dashboardGo to the ServiceWatch dashboard list screen
MoreDisplay additional tasks for managing charts
graph areaData graph collected during the period applied to the chart
  • Available graph
    • Average CPU Usage (Percent)
    • Average Disk Reads (Bytes)
    • Average Disk Writes (Bytes)
    • Summary Disk Read Requests (Count)
    • Summary Disk Write Requests (Count)
    • Maximum Network In (Bytes)
    • Maximum Network Out (Bytes)
    • Summary Instance State (Any)
  • When the mouse cursor is placed on the graph, a popup displays the time, data value, and metric data information for that point
  • You can drag the mouse to zoom in on a specific area of the graph
  • Clicking a label shown in the legend displays detailed information about that legend in a popup
Table. Virtual Server Monitoring Tab Chart Items
Reference
  • The metrics provided by basic monitoring are data collected at 5‑minute intervals.
  • For detailed information about the ServiceWatch metrics of Virtual Server, see Virtual Server의 ServiceWatch 지표.

Tag

On the Virtual Server List page, you can view the tag information of the selected resource, and you can add, modify, or delete it.

CategoryDetailed description
Tag listTag list
  • You can view the Key and Value information of the tag
  • Up to 50 tags can be added per resource
  • When entering a tag, you can search and select from the list of previously created Keys and Values
Table. Virtual Server Tag Tab Items

Job History

On the Virtual Server List page, you can view the operation history of the selected resource.

CategoryDetailed description
Task History ListResource Change History
  • You can view the operation details, operation time, resource type, resource name, operation result, and operator information
  • Operation History List Click the relevant resource in the list. Operation History Details popup window opens.
Table. Virtual Server operation history tab detailed information items

Control Virtual Server Operation

If you need to control the operation of a created Virtual Server resource, you can perform the task from the Virtual Server List or Virtual Server Details page. You can start, stop, and restart a running server.

Virtual Server Getting Started

You can start a Virtual Server that is shut off. To start the Virtual Server, follow the steps below.

  1. Click the All Services > Compute > Virtual Server menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. On the Service Home page, click the Virtual Server menu. 2. Go to the Virtual Server List page.
  3. Virtual Server List page, click the resource to start among the stopped (Shutoff) servers, and navigate to the Virtual Server Details page.
    • On the Virtual Server List page, you can Start each resource using the right More button.
    • After selecting multiple servers with checkboxes, you can control multiple servers simultaneously via the Start button at the top.
  4. On the Virtual Server Details page, click the Start button at the top to start the server. 4. Check the status of the changed server in the Status Indicator item.
    • When the Virtual Server start is complete, the server status changes from Shutoff to Active.
    • For detailed information about the Virtual Server status, please refer to Virtual Server 상세 정보 확인하기.

Stop Virtual Server

You can stop a Virtual Server that is running (Active). To stop the Virtual Server, follow these steps.

  1. All Services > Compute > Virtual Server Click the menu. 1. Go to the Service Home page of the Virtual Server.
  2. On the Service Home page, click the Virtual Server menu. 2. Virtual Server List Navigate to the page.
  3. Virtual Server List page, click the resource to stop among the servers that are active, and navigate to the Virtual Server Details page.
    • On the Virtual Server List page, you can Stop each resource using the right More button.
    • After selecting multiple servers with checkboxes, you can control multiple servers simultaneously using the Stop button at the top.
  4. On the Virtual Server Details page, click the Stop button at the top to start the server. 4. Check the status of the changed server in the Status Indicator item.
    • When the Virtual Server shutdown is complete, the server status changes from Active to Shutoff.
    • For detailed information about the Virtual Server status, please refer to Virtual Server 상세 정보 확인하기.

Restart Virtual Server

You can restart the created Virtual Server. To restart the Virtual Server, follow these steps.

  1. All Services > Compute > Virtual Server Click the menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. On the Service Home page, click the Virtual Server menu. 2. Go to the Virtual Server List page.
  3. Virtual Server List page, click the resource to restart, and navigate to the Virtual Server Details page.
    • Virtual Server List page allows you to restart each resource via the right More button.
    • After selecting multiple servers with checkboxes, you can control multiple servers simultaneously via the Restart button at the top.
  4. On the Virtual Server Details page, click the Restart button at the top to start the server. 4. Check the status of the changed server in the Status Indicator item.
    • During a Virtual Server restart, the server status passes through Rebooting and finally changes to Active.
    • For detailed information about the Virtual Server status, please refer to Virtual Server 상세 정보 확인하기.

Managing Virtual Server Resources

If you need server control and management functions for the created Virtual Server resources, you can perform tasks from the Virtual Server List or Virtual Server Details page.

Create Image

You can create an image of a running Virtual Server.

Reference

This document provides instructions on how to create a user image on a running Virtual Server.

  • Virtual Server List or Virtual Server Details page, click the Create Image button to generate a user Image.
  • Refer to Image 상세 가이드 내 Image 생성하기 for the method of uploading a user’s owned Image file to create an Image.

To create a Virtual Server Image, follow the steps below.

  1. All Services > Compute > Virtual Server Click the menu. 1. Navigate to the Service Home page of the Virtual Server.

  2. On the Service Home page, click the Virtual Server menu. 2. Go to the Virtual Server List page.

  3. On the Virtual Server List page, click the resource to create an Image. 3. Navigate to the Virtual Server Details page.

  4. On the Virtual Server Details page, click the Create Image button. 4. Go to the Image creation page.

    • Enter the required information in the Service Information Input area.
      Category
      Required
      Detailed description
      Image nameEssentialthe name of the Image to be created
      • Enter within 200 characters using English letters, numbers, spaces, and special characters(-, _)
      Table. Image service information input items
  5. Check the input information and click the Done button.

    • Once creation is complete, check the created resource on the All Services > Compute > Virtual Server > Image List page.
information
  • When an Image is created, the generated Image is stored in the Object Storage used as internal storage. * Therefore, usage fees will be charged for Image storage.
  • Since the file system of an Image created from an Active Virtual Server cannot be guaranteed to be intact, it is recommended to stop the server before creating the Image.

Edit server type

You can modify the server type of a Virtual Server.

Reference
For the configurable server types provided by Virtual Server, refer to Virtual Server 서버 타입.

To modify the server type of a Virtual Server, follow these steps.

  1. All Services > Compute > Virtual Server Click the menu. 1. Go to the Service Home page of the Virtual Server.
  2. On the Service Home page, click the Virtual Server menu. 2. Go to the Virtual Server List page.
  3. On the Virtual Server List page, click the resource to control its operation. 3. Navigate to the Virtual Server Details page.
  4. On the Virtual Server Details page, check the server status and click the server type Edit button. 4. Edit Server Type popup window opens.
  5. Edit Server Type In the popup window, after changing the server type, click the Confirm button.
    • If you modify the Virtual Server’s server type, the Virtual Server status changes to a state related to performing a resize.
    • For detailed information about the Virtual Server status, please refer to Virtual Server 상세 정보 확인하기.
Reference
If you change the Virtual Server’s server type, monitoring performance metric data may not be collected properly for a short period. In the next collection cycle (1 minute), normal performance metrics will be collected.

Change IP

Please refer to IP 변경하기 for instructions on changing the IP.

Caution
  • If you proceed with changing the IP, you will no longer be able to communicate using that IP, and you cannot cancel the IP change while it is in progress.
  • The server is rebooted to apply the changed IP.
  • If the server is running the Load Balancer service, you must delete the old IP from the LB server group and directly add the new IP as a member of the LB server group.
  • Servers using Public NAT/Private NAT must disable and reconfigure Public NAT/Private NAT after changing the IP.
    • If you are using Public NAT/Private NAT, first disable the use of Public NAT/Private NAT, complete the IP change, and then reconfigure.
    • Whether to use Public NAT/Private NAT can be changed by clicking the Edit button of Public NAT IP/Private NAT IP on the Virtual Server Details page.
  • Public NAT IP can only be modified when there is one server, the VPC is connected to an Internet Gateway, and the regular subnet is public.

Enable detailed monitoring for ServiceWatch

By default, Virtual Server is linked with ServiceWatch for basic monitoring. You can enable detailed monitoring as needed to identify operational issues more quickly and take action. For detailed information about ServiceWatch, refer to the ServiceWatch 개요.

Caution
Basic monitoring is provided for free, but enabling detailed monitoring incurs additional charges. Please note the usage.

To enable detailed ServiceWatch monitoring for a Virtual Server, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. On the Service Home page, click the Virtual Server menu. 2. Navigate to the Virtual Server List page.
  3. On the Virtual Server List page, click the resource to enable detailed ServiceWatch monitoring. 3. Navigate to the Virtual Server Details page.
  4. Click the Edit button for ServiceWatch detailed monitoring on the Virtual Server Details page. 4. ServiceWatch Detailed Monitoring Edit Navigate to the popup window.
  5. ServiceWatch Detailed Monitoring Edit In the popup window, after selecting Enable, review the instructions and click the Confirm button.
  6. On the Virtual Server Details page, view the detailed ServiceWatch monitoring items.

Disable detailed monitoring for ServiceWatch

Caution
Disabling detailed monitoring is required for cost efficiency. Maintain detailed monitoring only when absolutely necessary, and disable detailed monitoring otherwise.

To disable detailed ServiceWatch monitoring of a Virtual Server, follow these steps.

  1. All Services > Compute > Virtual Server Click the menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. On the Service Home page, click the Virtual Server menu. 2. Navigate to the Virtual Server List page.
  3. Virtual Server list on the page, click the resource to disable ServiceWatch detailed monitoring. 3. Navigate to the Virtual Server Details page.
  4. On the Virtual Server Details page, click the Edit button for ServiceWatch detailed monitoring. 4. ServiceWatch Detailed Monitoring Edit Navigate to the popup window.
  5. ServiceWatch Detailed Monitoring Edit In the popup window, after deselecting Enable, review the guidance message and click the Confirm button.
  6. On the Virtual Server Details page, check the ServiceWatch detailed monitoring items.

Virtual Server management add‑on features

For Virtual Server management, you can view console logs, generate dumps, and perform rebuilds. To view the console log, generate a dump, and rebuild a Virtual Server, follow these steps.

Check console log

You can view the current console log of the Virtual Server.

To view the console log of a Virtual Server, follow these steps.

  1. All Services > Compute > Virtual Server Click the menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. On the Service Home page, click the Virtual Server menu. 2. Go to the Virtual Server List page.
  3. Virtual Server List On the page, click the resource to view the console log. 3. Navigate to the Virtual Server Details page.
  4. Virtual Server Details on the page, click the Console Log button. 4. Console Log navigates to the popup window.
  5. Console Log Check the console log output in the popup window.

Create Dump

To create a Dump file of the Virtual Server, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. On the Service Home page, click the Virtual Server menu. 2. Navigate to the Virtual Server List page.
  3. On the Virtual Server List page, click the resource to view detailed information. 3. Navigate to the Virtual Server Details page.
  4. Click the Create Dump button on the Virtual Server Details page.
    • The dump file is created inside the Virtual Server.

Execute Rebuild

You can delete the OS data and configuration of the existing Virtual Server and rebuild it on a new server.

To perform a Rebuild of a Virtual Server, follow these steps.

  1. All Services > Compute > Virtual Server Click the menu. 1. Go to the Service Home page of the Virtual Server.
  2. On the Service Home page, click the Virtual Server menu. 2. Go to the Virtual Server List page.
  3. On the Virtual Server List page, click the resource to perform Rebuild. 3. Navigate to the Virtual Server Details page.
  4. On the Virtual Server Details page, click the Rebuild button.
    • During a Virtual Server Rebuild, the server status changes to Rebuilding, and when the Rebuild is complete, it returns to its pre‑Rebuild state.
    • For detailed information about the Virtual Server status, please refer to Virtual Server 상세 정보 확인하기.

Terminate Virtual Server

If you terminate an unused Virtual Server, you can reduce operating costs. However, if you terminate a Virtual Server, the running service may be stopped immediately, so you should proceed with the termination only after fully considering the impact of service interruption.

Caution
Please be aware that data cannot be recovered after terminating the service.

To cancel a Virtual Server, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. On the Service Home page, click the Virtual Server menu. 2. Virtual Server List page is accessed.
  3. On the Virtual Server List page, select the resource to terminate, and click the Terminate Service button.
    • The termination of attached storage depends on the Delete on termination setting; see 해지 제약 사항.
  4. If the termination is complete, check on the Virtual Server list page whether the resource has been terminated.

Termination constraints

When a Virtual Server termination request cannot be processed, a popup window will provide guidance. Please refer to the case below.

Cancellation not allowed
  • If File Storage is connected: Disconnect the File Storage first.
  • If the LB server group is connected: First, disconnect the LB server group pool connection.
  • When Lock is set: After changing the Lock setting to disabled, try again.
  • If the Auto-Scaling Group attached to the Virtual Server is not In Service: After changing the status of the attached Auto-Scaling Group, try again.

Termination of attached storage varies depending on the Delete on termination setting; please refer to it.

Delete on termination: delete per configuration
  • Whether the volume is deleted also depends on the Delete on termination setting.
    • Delete on termination when not set: Even if you terminate the Virtual Server, the volume will not be deleted.
    • Delete on termination when set: If the Virtual Server is terminated, the volume will be deleted.
  • Volumes that have a snapshot will not be deleted even if Delete on termination is set.
  • A multi-attach volume can be deleted only when the server being deleted is the last remaining server attached to the volume.

1.2.1 - Image

Users can create the service by entering the required information for the Image service within the Virtual Server service and selecting detailed options through the Samsung Cloud Platform Console.

Create Image

You can create and use the Image service while using the Virtual Server service in the Samsung Cloud Platform Console.

To create an Image, follow the steps below.

  1. All Services > Compute > Virtual Server Click the menu. 1. Navigate to the Service Home page of the Virtual Server.

  2. On the Service Home page, click the Image menu. 2. Go to the Image list page.

  3. Click the Create Image button on the Image List page. 3. Navigate to the Image creation page.

    • Enter or select the required information in the Service Information Input area.
      Category
      required status
      Detailed description
      Image nameRequiredthe name of the Image to create
      • Enter within 255 characters using English letters, numbers, spaces, and special characters(-, _)
      Image file > URLRequiredAfter uploading an image file to Object Storage, enter the URL
      • The URL can be copied on the Object Storage Details page
      • The bucket in Object Storage where the image file is uploaded must exist in the same zone as the server to be created
      • Image files can only be in the .qcow2 format
      • You must upload a secure image file to minimize security risks.
      OS typeRequiredOS type of the uploaded Image file
      • Select from Alma Linux, CentOS, Oracle Linux, RHEL, Rocky Linux, Ubuntu
      Minimum diskRequiredMinimum disk capacity (GB) of the Image to be created
      • 0~12,288GB, please enter a value
      Minimum RAMRequiredMinimum RAM size (GB) of the Image to be created
      • 0~2,097,151GB Enter a value
      VisibilityRequiredDisplay access permissions for the Image
      • Private: Only usable within the account
      • Shared: Can be shared across accounts
      ProtectedSelectionSelect whether Image deletion is prohibited
      • If you check Use, it prevents accidental deletion of the Image
      • This setting cannot be changed after the Image is created
      Table. Image service information input items
    • Additional Information Input in this area, enter or select the required information.
      Category
      required status
      Detailed description
      tagSelectionAdd Tag
      • Up to 50 per resource can be added
      • After clicking the Add Tag button, enter or select Key, Value values
      Table. Image additional information input fields
  4. Summary Check the detailed information and estimated billing amount generated in the panel, and click the Create button.

    • When creation is complete, check the created resources on the Image List page.

Check Image detailed information

The Image service allows you to view and edit the full resource list and detailed information. Image Details page consists of Details, Tags, Activity Log tabs.

To view detailed information of the Image service, follow the steps below.

  1. All Services > Compute > Virtual Server Click the menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. On the Service Home page, click the Image menu. 2. Go to the Image list page.
  3. On the Image list page, click the resource to view detailed information. 3. Navigate to the Image Details page.
    • Image Details page displays status information and additional feature information, and consists of Details, Tags, Activity Log tabs.
      CategoryDetailed description
      Image statusStatus of the Image created by the user
      • Active: Available state
      • Queued: When an Image creation request is made, the Image is uploaded and is pending processing
      • Importing: When an Image creation request is made, the Image is uploaded and is being processed
      Create shared ImageCreate an Image to share with another Account
      • Creation is possible only when the Image’s Visibility is private and the Image contains snapshot information
      • Creation is not allowed if the OS disk size is 500 GB or larger
      Share with another accountImage can be shared with another Account
      • If the Image’s Visibility is Shared, it can be shared with another Account
      • Create shared Image or displayed only on Images created by uploading a qcow2 file
      Delete ImageDelete Image button
      • Deleting the Image cannot be undone
      Table. Image status information and additional functions

Detailed Information

On the Image list page, you can view detailed information of the selected resource and edit the information if needed.

CategoryDetailed description
serviceservice name
Resource TypeResource Type
SRNUnique resource ID in Samsung Cloud Platform
  • Image refers to Image SRN
Resource nameImage name
Resource IDImage’s unique resource ID
ConstructorUser who created the Image
Creation date and timeImage creation timestamp
ModifierUser who edited the image information
Modification date and timeDate and time the image information was modified
Image nameImage name
Minimum diskImage’s minimum disk capacity (GB)
  • If you need to modify the minimum disk, click the Edit button to set it
Minimum RAMMinimum RAM size (GB) of the Image
OS typeImage OS type
  • Alma Linux, CentOS, Oracle Linux, RHEL, Rocky Linux, SLES, Ubuntu
OS hash algorithmOS hash algorithm method
VisibilityDisplay access permissions for the Image
  • Private: Only usable within the account
  • Shared: Can be shared across accounts
ProtectedSelect whether Image deletion is prohibited
  • Use setting prevents accidental deletion of Image
Image sizeImage size
  • If the size of the generated Image is 1 GB or less, it is displayed as 1 GB
  • For Images created via snapshot, not displayed
Image typeClassification by Image creation method
  • Snapshot-Based: When the configuration of the currently used Virtual Server is created as an Image
  • Image-Based: When an Image is created by uploading a qcow2 extension file or by creating a shared Image
Image file URLWhen creating an Image, the URL of the Image file uploaded to Object Storage
  • Virtual Server detail page does not show Images created via the Image creation menu, but shows Images uploaded to Object Storage
Sharing statusCurrent status of sharing the Image with another Account
  • Approved Account ID: ID of the Account for which sharing is approved
  • Modification timestamp: The time when sharing was requested to another Account, and if the sharing status later changes from Pending → Accepted, it is updated to that time
  • Status: Approved status
    • Accepted: sharing is approved and active
    • Pending: awaiting approval
  • Sharing stopped: sharing has been stopped
Table. Image detail information tab items

Tag

Image List page allows you to view the tag information of the selected resource, and to add, modify, or delete it.

CategoryDetailed description
Tag listTag list
  • You can view the Key and Value information of the tag
  • Up to 50 tags can be added per resource
  • When entering a tag, you can search and select from the list of previously created Keys and Values
Table. Image tag tab item

Job History

You can view the operation history of the selected resource on the Image list page.

CategoryDetailed description
Task History ListResource change history
  • Operation timestamp, resource ID, resource name, operation details, event topic, operation result, check operator information
Table. Image Work History Tab Detailed Information Items

Image resource management

Describes the control and management functions of the generated Image.

Create shared Image

Create an Image to share with another Account.

Notice
  • You can create a shared Image only when the Image’s Visibility is private and the Image contains snapshot information.
  • If the OS disk size is 500 GB or larger, a shared Image cannot be created.
  • Only one OS area disk volume is included in the imaging target for shared images. * Since the additional attached data volume is not included in the Image, if needed, copy the data to a separate volume and use the volume migration feature.

To create a shared Image, follow the steps below.

  1. Log in to the account to be shared and click the All Services > Compute > Virtual Server menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. On the Service Home page, click the Image menu. 2. Go to the Image list page.
  3. On the Image List page, click the Image to create a shared Image. 3. Navigate to the Image Details page.
  4. Click the Create Image for Sharing button. 4. A popup window opens to notify the creation of a shared Image.
  5. After reviewing the notification content, click the Complete button.

Share Image to another Account

Create an Image to share with another Account.

Notice
  • You can share only the Image created by uploading a .qcow2 extension file or by using Create Shared Image on the Image Details page with another Account.
  • The Visibility of the Image to be shared must be Shared.

To share an Image with another Account, follow these steps.

  1. Log in to the account to be shared and click the All Services > Compute > Virtual Server menu. 1. Navigate to the Service Home page of the Virtual Server.

  2. On the Service Home page, click the Image menu. 2. Go to the Image list page.

  3. Image List page, click the Image you want to share with another Account. 3. Go to the Image Details page.

  4. Share to another Account Click the button. 4. A popup window notifying image sharing opens.

  5. After reviewing the notification content, click the Confirm button. 5. Go to the Share Image with another Account page.

  6. Share Image with another Account on the page, enter Share Account ID, and click the Complete button. 6. A popup window notifying image sharing opens.

    Category
    required status
    Detailed description
    Image name-Name of the Image to share
    • Input not allowed
    Image ID-Image ID to share
    • Input not allowed
    Shared Account IDRequiredEnter another Account ID to share
    • English letters, numbers, and special characters- using them, enter within 64 characters
    Table. Image sharing items to another account

  7. After reviewing the notification content, click the Confirm button. 7. You can check the information from the sharing status on the Image Details page.

    • When the request is first made, the status is Pending, and when the receiving Account approves it, it changes to Accepted; if the approval is denied, it changes to Rejected.

Receive an Image shared from another Account

To receive an Image shared from another Account, follow these steps.

  1. Log in to the account to be shared and click the All Services > Compute > Virtual Server menu. 1. Navigate to the Service Home page of the Virtual Server.

  2. On the Service Home page, click the Image menu. 2. Go to the Image list page.

  3. On the Image List page, click the More > Image Share Request List button. 3. Image sharing request list The popup window opens.

  4. Image Sharing Request List In the popup window, click the Approve or Reject button for the Image to be shared.

    CategoryDetailed description
    Image nameName of shared Image
    OS typeOS type of shared Image
    Owner Account IDOwner Account ID of the shared Image
    Creation date and timeCreation time of shared Image
    ApprovalApprove the shared image
    RejectReject the shared image
    Table. Image sharing request list item

  5. After reviewing the notification content, click the Confirm button. 5. You can view the shared Image from the Image list.

Delete Image

You can delete unused Image. However, since an Image cannot be recovered once deleted, you should fully consider the impact before proceeding with the deletion.

Caution
Please be aware that data cannot be recovered after deleting the service.

To delete the Image, follow these steps.

  1. All Services > Compute > Virtual Server Click the menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. On the Service Home page, click the Image menu. 2. Go to the Image list page.
  3. On the Image list page, select the resource to delete, and click the Delete button.
    • On the Image List page, select multiple Image check boxes and click the Delete button at the top of the resource list.
  4. After the deletion is complete, verify on the Image List page that the resource has been removed.

1.2.2 - Keypair

Users can create the service by entering the required information for the Keypair within the Virtual Server service and selecting detailed options through the Samsung Cloud Platform Console.

Creating a Keypair

In the Samsung Cloud Platform Console, you can create and use the Keypair service while using the Virtual Server service.

To create a keypair, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. Navigate to the Service Home page of Virtual Server.

  2. On the Service Home page, click the Keypair menu. You will be taken to the Keypair List page.

  3. On the Keypair List page, click the Create Keypair button. You will be taken to the Create Keypair page.

    • Enter the required information in the Service Information Input area.
      Category
      Required status
      Detailed description
      Keypair nameRequiredName of the Keypair to create
      • Enter within 255 characters using English letters, numbers, spaces, and special characters(-, _)
      Keypair typeRequiredssh
      Table. Keypair service information input fields
    • In the Additional Information Input area, enter or select the required information.
      Category
      Required
      Detailed description
      tagSelectionAdd Tag
      • Up to 50 can be added per resource
      • After clicking the Add Tag button, enter or select Key, Value values
      Table. Keypair additional information input fields
  4. Check the input information and click the Create button.

    • After creation is complete, check the created resources on the Keypair list page.
Caution
  • After creation is complete, you can download the Key only once for the first time. Since reissuance is not possible, ensure that it has been downloaded.
  • Store the downloaded Private Key in a safe place.

View detailed information of the Keypair

The Keypair service allows you to view and edit the full list of resources and detailed information. On the Keypair Details page, it consists of Details, Tags, Activity Log tabs.

To view detailed information about a keypair, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. Navigate to the Service Home page of Virtual Server.
  2. On the Service Home page, click the Keypair menu. You will be taken to the Keypair List page.
  3. On the Keypair List page, click the resource to view detailed information. You will be taken to the Keypair Details page.
    • Keypair Details page displays status information and additional feature information, and consists of Details, Tags, Activity Log tabs.

Detailed Information

On the Keypair List page, you can view detailed information of the selected resource and edit the information if needed.

CategoryDetailed description
serviceService name
Resource TypeResource Type
SRNUnique resource ID in Samsung Cloud Platform
  • In Keypair, it refers to the Keypair SRN
Resource nameKeypair name
Resource IDUnique resource ID of the keypair
ConstructorUser who created the keypair
Creation date and timeKeypair creation timestamp
editorUser who modified the keypair information
Modification dateKeypair information modification timestamp
Keypair nameKeypair name
FingerprintA unique value for identifying the key
User IDUser ID of the keypair creator
public keyPublic key information
Table. Keypair detailed information tab items

tag

On the Keypair List page, you can view the tag information of the selected resource and add, modify, or delete it.

CategoryDetailed description
Tag listTag list
  • You can view the Key and Value information of the tag
  • Up to 50 tags can be added per resource
  • When entering a tag, you can search and select from the list of previously created Keys and Values
Table. Keypair Tag Tab Items

Job History

On the Keypair list page, you can view the operation history of the selected resource.

CategoryDetailed description
Task History ListResource Change History
  • Operation Time, Resource ID, Resource Name, Operation Details, Event Topic, Operation Result, Verify Operator Information
Table. Keypair operation history tab detailed information items

Keypair Resource Management

Describes the control and management functions of a keypair.

Get public key

To retrieve the public key, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. Go to the Service Home page of Virtual Server.

  2. On the Service Home page, click the Keypair menu. You will be taken to the Keypair List page.

  3. On the Keypair List page, click the More button at the top and then click the Import Public Key button. You will be taken to the Import Public Key page.

    • Required Information Input area, please enter or select the required information.
      Category
      Required
      Detailed description
      Keypair nameRequiredKeypair name to create
      Keypair typeRequiredssh
      public keyRequiredPublic Key Input
      • Load File: Select the Attach File button to attach the public key file
        • Only files with the .pem extension can be attached
      • Public Key Input: Paste the copied public key value
        • Keypair Details page allows copying the public key value
      Table. Required input fields for retrieving the public key
  4. Review the entered information and click the Complete button.

    • When creation is complete, check the created resource on the Keypair list page.
Notice
Even if you generate a new Keypair using Get Public Key, the Keypair cannot be re-downloaded. Use the Keypair that was issued at the time of initial creation.

Delete Keypair

You can delete unused Keypairs. However, once a Keypair is deleted it cannot be recovered, so please review the impact thoroughly beforehand before proceeding with deletion.

Caution
Please note that data cannot be recovered after deleting the service.

To delete a keypair, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. Go to the Service Home page of Virtual Server.
  2. On the Service Home page, click the Keypair menu. You will be taken to the Keypair List page.
  3. On the Keypair List page, select the resource to delete, and click the Delete button.
    • On the Keypair List page, select multiple Keypair check boxes and click the Delete button at the top of the resource list.
  4. After the deletion is complete, verify on the Keypair List page that the resource has been deleted.

1.2.3 - Server Group

Users can create the service by entering the required information for a Server Group within the Virtual Server service and selecting detailed options through the Samsung Cloud Platform Console.

Create Server Group

In the Samsung Cloud Platform Console, you can create and use the Server Group service while using the Virtual Server service.

To create a Server Group, follow the steps below.

  1. All Services > Compute > Virtual Server Click the menu. 1. Navigate to the Service Home page of the Virtual Server.

  2. On the Server Group page, click the Server Group menu. 2. Navigate to the Server Group List page.

  3. On the Server Group List page, click the Create Server Group button. 3. Server Group creation navigate to the page.

    • Enter or select the required information in the Service Information Input area.
      Category
      required status
      Detailed description
      Server Group nameRequiredEnter the name of the Server Group to be created
      • using English letters, numbers, spaces, and special characters (-, _) within 255 characters
      PolicyRequiredSelect the placement policy for Virtual Servers belonging to the same Server Group
      • Anti-Affinity(Distributed placement): A policy that places servers belonging to a Server Group on different racks and hosts as much as possible
      • Affinity(Proximity placement): A policy that places servers belonging to a Server Group close together within the same rack and host as much as possible
      • Partition(Virtual Server and Block Storage distributed placement): A policy that places Virtual Servers and their associated Block Storage from a Server Group into different distribution units (Partitions)
        • Up to three distributions are possible, and the setting can be applied when the number of possible distributions is two or more to ensure resource availability
        • Display the Partition number alongside to clearly indicate which Partition each Virtual Server and its associated Block Storage belong to
        • Partition numbers are assigned based on the Partition Size (up to 3) configured for the Server Group
      Table. Server Group Service Information Input Items
      Reference
      • Virtual Servers that belong to the same Server Group are placed on a Best Effort basis according to policy, but this is not strictly guaranteed.
    • Enter or select the required information in the Additional Information Input area.
      Category
      required status
      Detailed description
      tagSelectionAdd Tag
      • Up to 50 per resource can be added
      • After clicking the Add Tag button, input or select Key, Value values
      Table. Server Group additional information input items
  4. Check the input information and click the Generate button.

  5. When the popup notifying creation opens, click the Confirm button.

    • When creation is complete, check the created resources on the Server Group List page.

Check Server Group detailed information

The Server Group service lets you view and modify the complete resource list and detailed information. Server Group Details page consists of Details, Tags, Activity Log tabs.

To view detailed information about the Server Group, follow the steps below.

  1. All Services > Compute > Virtual Server Click the menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. On the Service Home page, click the Server Group menu. 2. Go to the Server Group List page.
  3. On the Server Group list page, click the resource to view detailed information. 3. Navigate to the Server Group Details page.
    • The Server Group Details page displays status information and additional feature information, and consists of Details, Tags, Operation History tabs.

Detailed Information

Server Group List page allows you to view detailed information of the selected resource and, if necessary, edit the information.

CategoryDetailed description
serviceservice name
Resource TypeResource Type
SRNUnique resource ID in Samsung Cloud Platform
  • In Server Group, it refers to the Server Group SRN
Resource nameServer Group name
Resource IDServer Group’s unique resource ID
ConstructorUser who created the Server Group
Creation date and timeServer Group creation date and time
Server Group nameServer Group name
PolicyAnti-Affinity(distributed placement), Affinity(proximal placement), Partition(distributed placement of Virtual Server and Block Storage)
Partition SizePartition Size
Server Group memberServer name and Partition information of Virtual Servers belonging to the Server Group
  • Members cannot be modified after the initial Virtual Server is created
  • Anti-Affinity (distributed placement), Affinity (proximate placement) policy: Since this policy defines only the relative placement relationship between Virtual Servers, the Samsung Cloud Platform Console provides only the list of Virtual Servers belonging to the policy
  • Partition (distributed placement of Virtual Server and Block Storage) policy: The Partition number is displayed together so that you can clearly see which Partition the Virtual Server and its associated Block Storage belong to
    • Partition numbers are assigned based on the Partition Size (up to 3) set for the Server Group
Table. Server Group detailed information tab items

Tag

Server Group list page lets you view the tag information of the selected resource, and add, modify, or delete it.

CategoryDetailed description
Tag listTag list
  • You can view the Key and Value information of the tag
  • Up to 50 tags can be added per resource
  • When entering a tag, you can search and select from the list of previously created Keys and Values
Table. Server Group Tag Tab Items

Job History

Server Group List page allows you to view the operation history of the selected resource.

CategoryDetailed description
Task History ListResource change history
  • Operation timestamp, resource ID, resource name, operation details, event topic, operation result, check operator information
Table. Server Group operation history tab detailed information items

Delete Server Group

You can delete unused Server Groups. However, since a Server Group cannot be recovered once deleted, please review the impact thoroughly in advance before proceeding with deletion.

Caution
Please be aware that data cannot be recovered after deleting the service.

To delete a Server Group, follow these steps.

  1. All Services > Compute > Virtual Server menu, click. 1. Navigate to the Service Home page of the Virtual Server.
  2. On the Service Home page, click the Server Group menu. 2. Go to the Server Group List page.
  3. On the Server Group List page, select the resource to delete, and click the Delete button.
    • On the Server Group List page, select multiple Server Group check boxes, and click the Delete button at the top of the resource list.
  4. After deletion is complete, check on the Server Group List page whether the resource has been deleted.

1.2.4 - Change IP

You can change the Virtual Server’s IP and add a network port to the Virtual Server to configure the IP.

Change IP

You can change the IP of the Virtual Server.

Caution
  • If you proceed with changing the IP, you will no longer be able to communicate using that IP, and you cannot cancel the IP change while it is in progress.
  • The server is rebooted to apply the changed IP.
  • If the server is running the Load Balancer service, you must delete the old IP from the LB server group and directly add the new IP as a member of the LB server group.
  • Servers using Public NAT/Private NAT must disable and reconfigure Public NAT/Private NAT after changing the IP.
    • If you are using Public NAT/Private NAT, first disable the use of Public NAT/Private NAT, complete the IP change, and then reconfigure.
    • You can change the usage of Public NAT/Private NAT by clicking the Edit button of Public NAT IP/Private NAT IP on the Virtual Server Details page.
  • Public NAT IP can only be modified when there is one server, the VPC is connected to an Internet Gateway, and the regular subnet is public.

To change the IP, follow these steps.

  1. All Services > Compute > Virtual Server Click the menu. 1. Go to the Service Home page of the Virtual Server.
  2. Click the Virtual Server menu on the Service Home page. 2. Go to the Virtual Server List page.
  3. On the Virtual Server list page, click the resource whose IP you want to change. 3. Virtual Server Details navigate to the page.
  4. On the Virtual Server Details page, click the Edit button for the IP entry you want to change. 4. IP edit The popup window opens.
  5. Edit IP in the popup window, after selecting Subnet, enter the IP to change.
  6. When the setup is complete, click the Confirm button.
  7. When the popup notifying IP modification opens, click the Confirm button.

Configure IP on the server after adding a network port

If you create a Virtual Server with Ubuntu Linux, after adding a network port in Samsung Cloud Platform, you also need to configure the IP on the server.

  1. As the root user on the Virtual Server’s OS, use the ip command to verify the assigned network interface name.

    Color mode
    ip a
    ip a
    코드블록. ip 명령어 - 네트워크 인터페이스 확인 명령어

    • If an added interface is present, the following result is displayed.
    Color mode
    [root@scp-test-vm-01 ~] # ip a
    3: ens7: <BROADCAST,MULTICAST> mtu 9000 qdisc noop state DOWN group default qlen 1000
        link/ether fa:16:3e:98:b6:64 brd ff:ff:ff:ff:ff:ff
        altname enp0s7
    [root@scp-test-vm-01 ~] # ip a
    3: ens7: <BROADCAST,MULTICAST> mtu 9000 qdisc noop state DOWN group default qlen 1000
        link/ether fa:16:3e:98:b6:64 brd ff:ff:ff:ff:ff:ff
        altname enp0s7
    코드블록. ip 명령어 - 네트워크 인터페이스 확인 결과
  2. Open the /etc/netplan/50-cloud-init.yaml file using a text editor (e.g., vim).

  3. Add the following content to the /etc/netplan/50-cloud-init.yaml file and save it.

    Color mode
    network:
    version: 2
    ethernets:
    ens7:
    match:
    macaddress: "fa:16:3e:98:b6:64"
          dhcp4: true
          set-name: "ens7"
          mtu: 9000
    network:
    version: 2
    ethernets:
    ens7:
    match:
    macaddress: "fa:16:3e:98:b6:64"
          dhcp4: true
          set-name: "ens7"
          mtu: 9000
    코드블록. YAML 파일 편집
Reference
Indentation is important in YAML files that configure netplan. When modifying a YAML file, please refer to the existing settings and be careful.
  1. Use the netplan command to set an IP on the added network DEVICE.

    Color mode
    netplan --debug apply
    netplan --debug apply
    코드블록. netplan 적용
  2. Use the ip command to verify that the IP is set correctly.

    Color mode
    [root@scp-test-vm-01 ~] # ip a
    3: ens7: <BROADCAST,MULTICAST> mtu 9000 qdisc noop state DOWN group default qlen 1000
        link/ether fa:16:3e:98:b6:64 brd ff:ff:ff:ff:ff:ff
        altname enp0s7
        inet 10.10.10.10/24 metric 100 brd 10.10.10.255 scope global dynamic ens7
           valid_lft 43197sec preferred_lft 43197sec
        inet6 fe80::f816:3eff:fe0a:96bf/64 scope link
           valid_lft forever preferred_lft forever
    [root@scp-test-vm-01 ~] # ip a
    3: ens7: <BROADCAST,MULTICAST> mtu 9000 qdisc noop state DOWN group default qlen 1000
        link/ether fa:16:3e:98:b6:64 brd ff:ff:ff:ff:ff:ff
        altname enp0s7
        inet 10.10.10.10/24 metric 100 brd 10.10.10.255 scope global dynamic ens7
           valid_lft 43197sec preferred_lft 43197sec
        inet6 fe80::f816:3eff:fe0a:96bf/64 scope link
           valid_lft forever preferred_lft forever
    코드블록. IP 설정 확인

1.2.5 - Configure Linux NTP

Users who create a Virtual Server with Rocky Linux or Oracle Linux via the Samsung Cloud Platform Console need additional configuration for time synchronization (NTP:Network Time Protocol). For other standard Linux images (RHEL, Alma Linux, Ubuntu), NTP is already configured, so no extra setup is required.

Installing NTP Daemon

You can configure NTP by installing the chrony daemon. To install the chrony daemon, follow the steps below.

Reference
For detailed information about chrony, see the chronyc page.
  1. Check whether the chrony package is installed by using the dnf command as the root user on the OS of the Virtual Server.

    Color mode
    dnf list chrony
    dnf list chrony
    Code block. dnf command - command to verify the chrony package installation

    • If the chrony package is installed, the following result is displayed.
Color mode
[root@scp-test-vm-01 ~] # dnf list chrony
Last metadata expiration check: 1:47:29 ago on Wed 19 Feb 2025 05:55:57 PM KST.
Installed Packages
chrony.x86_64                              3.5-1.0.1.el8                                              @anaconda
[root@scp-test-vm-01 ~] # dnf list chrony
Last metadata expiration check: 1:47:29 ago on Wed 19 Feb 2025 05:55:57 PM KST.
Installed Packages
chrony.x86_64                              3.5-1.0.1.el8                                              @anaconda
code block. dnf command - result of checking chrony package installation
  1. If the chrony package is not installed, install the chrony package using the dnf command.
    Color mode
    dnf install chrony -y
    dnf install chrony -y
    code block. dnf command - chrony package installation verification command

Configuring NTP Daemon

Reference
For detailed information about chrony, see the chronyc page.

To configure the chrony daemon, follow these steps.

  1. Open the /etc/chrony.conf file using a text editor (e.g., vim).
  2. Add the following content to the /etc/chrony.conf file and save it.
Color mode
server 198.19.0.54 iburst
server 198.19.0.54 iburst
Code block. Edit /etc/chrony.conf
  1. Use the systemctl command to configure the chrony daemon to start automatically.
Color mode
systemctl enable chronyd
systemctl enable chronyd
Code block. systemctl command - configure automatic start of chrony daemon
  1. Restart the chrony daemon using the systemctl command.

    Color mode
    systemctl restart chronyd
    systemctl restart chronyd
    code block. systemctl command - restart chrony daemon

  2. Run the chronyc sources command with the “v” option (display detailed information) to check the IP address of the configured NTP server and verify whether synchronization is in progress.

    Color mode
    chronyc sources -v
    chronyc sources -v
    Code block. chronyc sources command - check NTP synchronization

    • When you run the chronyc sources command, the following result is displayed.
Color mode
[root@scp-test-vm-01 ~] # chronyc sources -v

210 Number of sources = 1



  .-- Source mode   '^' = server,   '=' = peer,   '#' = local clock.

 / .- Source state     '*' = current synced,   '+' = combined ,   '-' = not combined,

| /    '?' = unreachable,   'x' = time may be in error,   '~' = time too variable.

|| .- xxxx [ yyyy ] +/- zzzz

||                      Reachability register (octal) -.                        |  xxxx = adjusted offset,

||                      Log2(Polling interval) --.          |                    |  yyyy = measured offset,

||                                                             |                   |  zzzz = estimated error.

||                                                        |       |                   

MS Name/IP address                  Stratum   Poll   Reach   LastRx   Last    sample

=========================================================================

^* 198.19.0.54                                 2      6     377      52      -129us[  -128us] +/-     14ms
[root@scp-test-vm-01 ~] # chronyc sources -v

210 Number of sources = 1



  .-- Source mode   '^' = server,   '=' = peer,   '#' = local clock.

 / .- Source state     '*' = current synced,   '+' = combined ,   '-' = not combined,

| /    '?' = unreachable,   'x' = time may be in error,   '~' = time too variable.

|| .- xxxx [ yyyy ] +/- zzzz

||                      Reachability register (octal) -.                        |  xxxx = adjusted offset,

||                      Log2(Polling interval) --.          |                    |  yyyy = measured offset,

||                                                             |                   |  zzzz = estimated error.

||                                                        |       |                   

MS Name/IP address                  Stratum   Poll   Reach   LastRx   Last    sample

=========================================================================

^* 198.19.0.54                                 2      6     377      52      -129us[  -128us] +/-     14ms
Code block. chronyc sources command - check NTP synchronization
  1. Run the chronyc tracking command to check the synchronization metrics.
    Color mode
    [root@scp-test-vm-01 ~] # chronyc tracking
    Reference ID        : A9FEA9FE (198.19.0.54)
    Stratum              : 3
    Ref time  (UTC)     : Wed  Feb  19  18:48:41  2025
    System time         : 0.000000039 seconds fast of NTP time
    Last offset            : -0.000084246 seconds
    RMS offset           : 0.000084246 seconds
    Frequency            : 21.667 ppm slow
    Residual freq        : +4.723 ppm
    Skew                  : 0.410 ppm
    Root delay           : 0.000564836 seconds
    Root dispersion     : 0.027399288 seconds
    Update interval      : 2.0 seconds
    Leap status           : Normal
    [root@scp-test-vm-01 ~] # chronyc tracking
    Reference ID        : A9FEA9FE (198.19.0.54)
    Stratum              : 3
    Ref time  (UTC)     : Wed  Feb  19  18:48:41  2025
    System time         : 0.000000039 seconds fast of NTP time
    Last offset            : -0.000084246 seconds
    RMS offset           : 0.000084246 seconds
    Frequency            : 21.667 ppm slow
    Residual freq        : +4.723 ppm
    Skew                  : 0.410 ppm
    Root delay           : 0.000564836 seconds
    Root dispersion     : 0.027399288 seconds
    Update interval      : 2.0 seconds
    Leap status           : Normal
    code block. chronyc tracking command - NTP synchronization metric

1.2.6 - Configure RHEL Repo and WKMS

information
  • If a user creates RHEL and Windows Server prior to August 2025 via the Samsung Cloud Platform Console, they need to modify the RHEL Repository and WKMS (Windows Key Management Service) settings.
  • The SCP RHEL Repository is a repository provided by SCP to support user environments such as VPC Private Subnets where external access is restricted.
    Since the SCP RHEL Repository synchronizes with each Region Local Repository according to an internal schedule, it is recommended to switch to an external public mirror site to apply the latest patches quickly.

RHEL Repository Configuration Guide

In Samsung Cloud Platform, when using RHEL, you can install and download the same packages as the official RHEL Repository by using the RHEL Repository provided by SCP.
SCP provides the latest repository for the given major version by default. To configure the RHEL repository, follow the steps below.

  1. As the root user on the Virtual Server, use the cat command to verify the /etc/yum.repos.d/scp.rhel8.repo or /etc/yum.repos.d/scp.rhel9.repo settings.

    Color mode
    cat /etc/yum.repos.d/scp.rhel8.repo
    cat /etc/yum.repos.d/scp.rhel8.repo
    Code block. Verify repo configuration (RHEL8)
    Color mode
    cat /etc/yum.repos.d/scp.rhel9.repo
    cat /etc/yum.repos.d/scp.rhel9.repo
    Code block. Verify repo configuration (RHEL9)

    • When checking the configuration file, the following result is displayed.
      Color mode
      [rhel-8-baseos]
      name=rhel-8-baseos
      gpgcheck=0
      enabled=1
      baseurl=http://scp-rhel8-ip/rhel/8/baseos
      [rhel-8-baseos-debug]
      name=rhel-8-baseos-debug
      gpgcheck=0
      enabled=1
      baseurl=http://scp-rhel8-ip/rhel/8/baseos-debug
      [rhel-8-appstream]
      name=rhel-8-appstream
      gpgcheck=0
      enabled=1
      baseurl=http://scp-rhel8-ip/rhel/8/appstream
      [rhel-8-baseos]
      name=rhel-8-baseos
      gpgcheck=0
      enabled=1
      baseurl=http://scp-rhel8-ip/rhel/8/baseos
      [rhel-8-baseos-debug]
      name=rhel-8-baseos-debug
      gpgcheck=0
      enabled=1
      baseurl=http://scp-rhel8-ip/rhel/8/baseos-debug
      [rhel-8-appstream]
      name=rhel-8-appstream
      gpgcheck=0
      enabled=1
      baseurl=http://scp-rhel8-ip/rhel/8/appstream
      Code block. Verify repo configuration (RHEL8)
      Color mode
      [rhel-9-for-x86_64-baseos-rpms]
      name=rhel-9-for-x86_64-baseos-rpms
      baseurl=http://scp-rhel9-ip/rhel/$releasever/x86_64/baseos
      gpgcheck=0
      enabled=1
      [rhel-9-for-x86_64-appstream-rpms]
      name=rhel-9-for-x86_64-appstream-rpms
      baseurl=http://scp-rhel9-ip/rhel/$releasever/x86_64/appstream
      gpgcheck=0
      enabled=1
      [codeready-builder-for-rhel-9-x86_64-rpms]
      name=codeready-builder-for-rhel-9-x86_64-rpms
      baseurl=http://scp-rhel9-ip/rhel/$releasever/x86_64/codeready-builder
      gpgcheck=0
      enabled=1
      [rhel-9-for-x86_64-highavailability-rpms]
      name=rhel-9-for-x86_64-highavailability-rpms
      baseurl=http://scp-rhel9-ip/rhel/$releasever/x86_64/ha
      gpgcheck=0
      enabled=1
      [rhel-9-for-x86_64-supplementary-rpms]
      name=rhel-9-for-x86_64-supplementary-rpms
      baseurl=http://scp-rhel9-ip/rhel/$releasever/x86_64/supplementary
      gpgcheck=0
      enabled=1
      [rhel-9-for-x86_64-baseos-rpms]
      name=rhel-9-for-x86_64-baseos-rpms
      baseurl=http://scp-rhel9-ip/rhel/$releasever/x86_64/baseos
      gpgcheck=0
      enabled=1
      [rhel-9-for-x86_64-appstream-rpms]
      name=rhel-9-for-x86_64-appstream-rpms
      baseurl=http://scp-rhel9-ip/rhel/$releasever/x86_64/appstream
      gpgcheck=0
      enabled=1
      [codeready-builder-for-rhel-9-x86_64-rpms]
      name=codeready-builder-for-rhel-9-x86_64-rpms
      baseurl=http://scp-rhel9-ip/rhel/$releasever/x86_64/codeready-builder
      gpgcheck=0
      enabled=1
      [rhel-9-for-x86_64-highavailability-rpms]
      name=rhel-9-for-x86_64-highavailability-rpms
      baseurl=http://scp-rhel9-ip/rhel/$releasever/x86_64/ha
      gpgcheck=0
      enabled=1
      [rhel-9-for-x86_64-supplementary-rpms]
      name=rhel-9-for-x86_64-supplementary-rpms
      baseurl=http://scp-rhel9-ip/rhel/$releasever/x86_64/supplementary
      gpgcheck=0
      enabled=1
      Code block. Check repo configuration (RHEL9)
  2. Use a text editor (e.g., vim) to open the /etc/hosts file.

  3. Modify the /etc/hosts file with the following content and save it.

    Color mode
    198.19.2.13 scp-rhel8-ip scp-rhel9-ip scp-rhel-ip
    198.19.2.13 scp-rhel8-ip scp-rhel9-ip scp-rhel-ip
    code block. Change /etc/hosts file settings

  4. Use the yum command to verify the RHEL repository connection configured on the server.

    Color mode
    yum repolist –v
    yum repolist –v
    Code block. Verify repository connection settings

    • If the RHEL repository is successfully connected, you can view the repository list.
      Color mode
      Repo-id            : rhel-8-appstream
      Repo-name          : rhel-8-appstream
      Repo-revision      : 1718903734
      Repo-updated       : Fri 21 Jun 2024 02:15:34 AM KST
      Repo-pkgs          : 38,260
      Repo-available-pkgs: 25,799
      Repo-size          : 122 G
      Repo-baseurl       : http://scp-rhel8-ip/rhel/8/appstream
      Repo-expire        : 172,800 second(s) (last: Thu 08 Aug 2024 07:27:57 AM KST)
      Repo-filename      : /etc/yum.repos.d/scp.rhel8.repo
      
      Repo-id            : rhel-8-baseos
      Repo-name          : rhel-8-baseos
      Repo-revision      : 1718029433
      Repo-updated       : Mon 10 Jun 2024 11:23:52 PM KST
      Repo-pkgs          : 17,487
      Repo-available-pkgs: 17,487
      Repo-size          : 32 G
      Repo-baseurl       : http://scp-rhel8-ip/rhel/8/baseos
      Repo-expire        : 172,800 second(s) (last: Thu 08 Aug 2024 07:27:57 AM KST)
      Repo-filename      : /etc/yum.repos.d/scp.rhel8.repo
      
      Repo-id            : rhel-8-baseos-debug
      Repo-name          : rhel-8-baseos-debug
      Repo-revision      : 1717662461
      Repo-updated       : Thu 06 Jun 2024 05:27:41 PM KST
      Repo-pkgs          : 17,078
      Repo-available-pkgs: 17,078
      Repo-size          : 100 G
      Repo-baseurl       : http://scp-rhel8-ip/rhel/8/baseos-debug
      Repo-expire        : 172,800 second(s) (last: Thu 08 Aug 2024 07:27:57 AM KST)
      Repo-filename      : /etc/yum.repos.d/scp.rhel8.repo
      Repo-id            : rhel-8-appstream
      Repo-name          : rhel-8-appstream
      Repo-revision      : 1718903734
      Repo-updated       : Fri 21 Jun 2024 02:15:34 AM KST
      Repo-pkgs          : 38,260
      Repo-available-pkgs: 25,799
      Repo-size          : 122 G
      Repo-baseurl       : http://scp-rhel8-ip/rhel/8/appstream
      Repo-expire        : 172,800 second(s) (last: Thu 08 Aug 2024 07:27:57 AM KST)
      Repo-filename      : /etc/yum.repos.d/scp.rhel8.repo
      
      Repo-id            : rhel-8-baseos
      Repo-name          : rhel-8-baseos
      Repo-revision      : 1718029433
      Repo-updated       : Mon 10 Jun 2024 11:23:52 PM KST
      Repo-pkgs          : 17,487
      Repo-available-pkgs: 17,487
      Repo-size          : 32 G
      Repo-baseurl       : http://scp-rhel8-ip/rhel/8/baseos
      Repo-expire        : 172,800 second(s) (last: Thu 08 Aug 2024 07:27:57 AM KST)
      Repo-filename      : /etc/yum.repos.d/scp.rhel8.repo
      
      Repo-id            : rhel-8-baseos-debug
      Repo-name          : rhel-8-baseos-debug
      Repo-revision      : 1717662461
      Repo-updated       : Thu 06 Jun 2024 05:27:41 PM KST
      Repo-pkgs          : 17,078
      Repo-available-pkgs: 17,078
      Repo-size          : 100 G
      Repo-baseurl       : http://scp-rhel8-ip/rhel/8/baseos-debug
      Repo-expire        : 172,800 second(s) (last: Thu 08 Aug 2024 07:27:57 AM KST)
      Repo-filename      : /etc/yum.repos.d/scp.rhel8.repo
      Code block. Check repository list

Windows Key Management Service Configuration Guide

In Samsung Cloud Platform, when using Windows Server, you can activate genuine licensing by utilizing the Key Management Service provided by SCP. Follow the steps below.

  1. Right-click the Windows Start icon, then run cmd from Windows PowerShell (Administrator) or the Windows Run menu.

  2. In Windows PowerShell (Administrator) or cmd, please register the KMS Server by running the command below.

    Color mode
    slmgr /skms 198.19.2.23:1688
    slmgr /skms 198.19.2.23:1688
    Code block. WKMS configuration

  3. After executing the KMS Server registration command, confirm the notification popup indicating successful registration, then click OK.

    Figure
    Figure. Verify WKMS settings
  4. In Windows PowerShell (Administrator) or cmd, run the command below to activate the product.

    Color mode
    slmgr /ato
    slmgr /ato
    Code block. Windows Server activation settings

  5. After confirming the notification popup that the product activation was completed successfully, click OK.

    Figure
    Figure. Verify Windows Server activation
  6. Please run the command below in Windows PowerShell (Administrator) or cmd to verify that activation has been completed.

    Color mode
    slmgr /dlv
    slmgr /dlv
    Code block. Verify Windows Server activation

  7. After confirming the notification popup that the product activation was completed successfully, click OK.

    Figure
    Figure. Windows Server genuine activation verification

1.2.7 - Install ServiceWatch Agent

Users can install the ServiceWatch Agent on a Virtual Server to collect custom metrics and logs.

Reference
Collecting custom metrics/logs via the ServiceWatch Agent is currently available only on Samsung Cloud Platform For Enterprise. It will also be available in other offerings in the future.
Caution
Metric collection through the ServiceWatch Agent is classified as custom metrics and, unlike the default metrics collected from each service, incurs charges; therefore, we recommend removing or disabling unnecessary metric collection settings.

ServiceWatch Agent

The agents that must be installed on a Virtual Server to collect ServiceWatch’s custom metrics and logs can be divided into two main types. It is the Prometheus Exporter and OpenTelemetry Collector.

CategoryDetailed description
Prometheus ExporterProvide metrics of a specific application or service in a format that Prometheus can scrape
  • Depending on the OS, you can use the Node Exporter for Linux servers and the Windows Exporter for Windows servers
Open Telemetry CollectorActs as a centralized collector that gathers telemetry data such as metrics and logs from distributed systems, processes (filtering, sampling, etc.), and exports them to various backends (e.g., Prometheus, Jaeger, Elasticsearch, etc.)
  • Exports data via the ServiceWatch Gateway so that ServiceWatch can collect metric and log data.
Table. Explanation of Prometheus Exporter and Open Telemetry Collector

Pre-configuration for Using ServiceWatch Agent

To use the ServiceWatch Agent, please refer to ServiceWatch Agent를 위한 사전 환경 설정 and prepare the prerequisite configuration.

Install Prometheus Exporter for Virtual Server (for Linux)

Install the Prometheus Exporter on a Linux server according to the steps below.

Node Exporter installation

Install Node Exporter according to the steps below.

  1. Node Exporter User Creation
  2. Node Exporter configuration

Create Node Exporter User

Create a dedicated user to securely isolate the Node Exporter process.

Color mode
sudo useradd --no-create-home --shell /bin/false node_exporter
sudo useradd --no-create-home --shell /bin/false node_exporter
code block. Node Exporter user creation command

Node Exporter configuration

  1. Download to install Node Exporter. This guide refers to the version below.
    • Download path: /tmp
    • Installed version: 1.7.0
      Color mode
      cd /tmp
      wget https://github.com/prometheus/node_exporter/releases/download/v1.7.0/node_exporter-1.7.0.linux-amd64.tar.gz
      cd /tmp
      wget https://github.com/prometheus/node_exporter/releases/download/v1.7.0/node_exporter-1.7.0.linux-amd64.tar.gz
      Code block. Node Exporter download command
Reference
The latest version of Node Exporter can be found at Node Exporter > Releases > Lastest, and specific versions of Node Exporter can be found at Node Exporter > Releases.
  1. Install the downloaded Node Exporter and grant permission to the executable file.

    Color mode
    cd /tmp
    sudo tar -xvf node_exporter-1.7.0.linux-amd64.tar.gz -C /usr/local/bin --strip-components=1 node_exporter-1.7.0.linux-amd64/node_exporter
    cd /tmp
    sudo tar -xvf node_exporter-1.7.0.linux-amd64.tar.gz -C /usr/local/bin --strip-components=1 node_exporter-1.7.0.linux-amd64/node_exporter
    code block. Node Exporter installation command
    Color mode
    sudo chown node_exporter:node_exporter /usr/local/bin/node_exporter
    sudo chown node_exporter:node_exporter /usr/local/bin/node_exporter
    Code block. Node Exporter permission setting command

  2. Create service file Configure Node Exporter to collect memory metrics (meminfo) or block storage metrics (filesystem).

    Color mode
    sudo vi /etc/systemd/system/node_exporter.service
    sudo vi /etc/systemd/system/node_exporter.service
    Code block. Command to open the Node Exporter service file
    Color mode
    [Unit]
    Description=Prometheus Node Exporter (meminfo only)
    Wants=network-online.target
    After=network-online.target
    
    [Service]
    User=node_exporter
    Group=node_exporter
    Type=simple
    ExecStart=/usr/local/bin/node_exporter \
      --collector.disable-defaults \    # disable default metrics
      --collector.meminfo \             # Enable memory metrics
      --collector.filesystem            # Block Storage filesystem metrics enable
    
    Restart=on-failure
    
    [Install]
    WantedBy=multi-user.target
    [Unit]
    Description=Prometheus Node Exporter (meminfo only)
    Wants=network-online.target
    After=network-online.target
    
    [Service]
    User=node_exporter
    Group=node_exporter
    Type=simple
    ExecStart=/usr/local/bin/node_exporter \
      --collector.disable-defaults \    # disable default metrics
      --collector.meminfo \             # Enable memory metrics
      --collector.filesystem            # Block Storage filesystem metrics enable
    
    Restart=on-failure
    
    [Install]
    WantedBy=multi-user.target
    code block. Result of opening the Node Exporter service file

Reference
  • The collector can be enabled or disabled using a flag.

    • –collector.{name}: Used to enable a specific metric.
    • –no-collector.{name}: Disables a specific metric.
    • To disable all default metrics and enable only a specific collector, you can use –collector.disable-defaults –collector.{name} ….
  • Below is a description of the main collector.

CollectorExplanationLabel
meminfoProvide memory statistics-
filesystemProvides file system statistics such as used disk space
  • device: the physical or virtual device path where the filesystem is located
    • Example: /dev/sda1
  • fstype: the type of the filesystem
    • Example: ext4, xfs, nfs, tmpfs
  • mountpoint: the path where the filesystem is mounted on the host OS. The most intuitive criterion for distinguishing disks
    • Example: /, /var/lib/docker, /mnt/data
Table: Description of major Node Exporter collectors
  • Node Exporter 지표 you can check the main metrics provided by Node Exporter and the Node Exporter Collector configuration method.
Reference
  • Detailed information about the collectable metrics and how to configure them can be found in Node Exporter > Collector.
  • The metrics available may vary depending on the version of the Node Exporter you are using. * Please refer to Node Exporter.
Caution
Since metric collection through the ServiceWatch Agent is classified as custom metrics and incurs charges unlike the default metrics, unnecessary metric collection must be removed or disabled to prevent excessive fees.
  1. Service activation and start Register the Node Exporter service and verify the registered service and configured metrics.
    Color mode
    sudo systemctl daemon-reload
    sudo systemctl enable --now node_exporter
    sudo systemctl daemon-reload
    sudo systemctl enable --now node_exporter
    Code block. Node Exporter service activation and start command
    Color mode
    sudo systemctl status node_exporter
    sudo systemctl status node_exporter
    Code block. Node Exporter service check command
    Color mode
    curl http://localhost:9100/metrics | grep node_memory
    curl http://localhost:9100/metrics | grep node_memory
    Code block. Node Exporter metric information check command
information
If you have completed the Node Exporter configuration, you must install the Open Telemetry Collector provided by ServiceWatch to finish setting up the ServiceWatch Agent.
For more details, see ServiceWatch > ServiceWatch Agent 사용하기.

Install Prometheus Exporter for Virtual Server (for Windows)

Install according to the steps below to use Prometheus Exporter on a Windows server.

Install Windows Exporter

Install the Windows Exporter following the steps below.

  1. Windows Exporter 설정

Windows Exporter configuration

  1. Download the installation file to install the Windows Exporter.
    • Download path: C:\Temp
    • Test version: 0.31.3
      Color mode
      $ mkdir /Temp
      $ Invoke-WebRequest -Uri "https://github.com/prometheus-community/windows_exporter/releases/download/v0.31.3/windows_exporter-0.31.3-amd64.exe" -OutFile "C:\Temp\windows_exporter-0.31.3-amd64.exe"
      $ mkdir /Temp
      $ Invoke-WebRequest -Uri "https://github.com/prometheus-community/windows_exporter/releases/download/v0.31.3/windows_exporter-0.31.3-amd64.exe" -OutFile "C:\Temp\windows_exporter-0.31.3-amd64.exe"
      code block. Download Windows Exporter
Reference
Windows Exporter version and installation files can be found at Windows Exporter > Releases.
  1. Windows Exporter execution test
    Windows Exporter enables all collectors by default, but to collect only the metrics you want, you need to enable the following collectors. The following is an example that activates a collector specified by the user.
    • Memory metric: memory
    • Block storage metric: local_disk
    • Host name: os
      Color mode
      $ cd C:\Temp
      $ .\windows_exporter-0.31.3-amd64.exe --collectors.enabled memory,logical_disk,os
      $ cd C:\Temp
      $ .\windows_exporter-0.31.3-amd64.exe --collectors.enabled memory,logical_disk,os
      code block. Windows Exporter execution test
Reference
  • The collector can be enabled using a flag.

    • –collectors.enabled “[defaults]" Metrics provided by default
    • –collector.enabled {name},{name},{name}…: Used to enable specific metrics.
  • Below is a description of the main collector.

CollectorExplanationLabel
memoryProvide memory statistics
logical_diskCollect performance and health metrics of logical disks (e.g., C:, D: drives) on the local system.
  • volume: the physical or virtual device path where the file system is located
    • Example: C:D:
Table. Description of major Windows Exporter collectors
  • Windows Exporter 지표 you can find the main metrics provided by Windows Exporter and the method to configure the Windows Exporter Collector.
Reference
  • Detailed information about the collectable metrics and how to configure them can be found in Windows Exporter > Collector.

  • The metrics that can be provided may vary depending on the version of the Windoes Exporter you are using. * Please refer to Windows Exporter.

Caution
Since metric collection through the ServiceWatch Agent is classified as custom metrics and incurs charges unlike the default metrics, unnecessary metric collection must be removed or disabled to prevent excessive fees.
  1. Service registration and verification Register the Windows Exporter service and verify the configured metrics.

    Color mode
    $ sc.exe create windows_exporter binPath= "C:\Temp\windows_exporter-0.31.3-amd64.exe --collectors.enabled memory,logical_disk,os" DisplayName= "Prometheus Windows Exporter" start= auto
    $ Start-Service windows_exporter
    $ sc.exe create windows_exporter binPath= "C:\Temp\windows_exporter-0.31.3-amd64.exe --collectors.enabled memory,logical_disk,os" DisplayName= "Prometheus Windows Exporter" start= auto
    $ Start-Service windows_exporter
    code block. Windows Exporter service registration
    Color mode
    # Service check
    $ Get-Service windows_exporter
    
    # Check metrics
    $ Invoke-WebRequest -Uri "http://localhost:9182/metrics" | Select-String memory
    # Service check
    $ Get-Service windows_exporter
    
    # Check metrics
    $ Invoke-WebRequest -Uri "http://localhost:9182/metrics" | Select-String memory
    Code block. Check Windows Exporter service

  2. Configuration file settings

    • You can set it to use a YAML configuration file by using the –config.file option.
      Color mode
      $ .\windows_exporter.exe --config.file=config.yml
      $ .\windows_exporter.exe --config.file="C:\Program Files\windows_exporter\config.yml" # If you use an absolute path, you must enclose the path in quotes.
      $ .\windows_exporter.exe --config.file=config.yml
      $ .\windows_exporter.exe --config.file="C:\Program Files\windows_exporter\config.yml" # If you use an absolute path, you must enclose the path in quotes.
      code block. Windows Exporter configuration file settings
      Color mode
      collectors:
        enabled: cpu,net,service
      collector:
        service:
          include: windows_exporter
      log:
        level: warn
      collectors:
        enabled: cpu,net,service
      collector:
        service:
          include: windows_exporter
      log:
        level: warn
      Code block. Example of a part of a Windows Exporter configuration file
    • Please refer to Windows Exporter > Official Example Configuration File.
      Color mode
      ---
      #Note this is not an exhaustive list of all configuration values
      collectors:
        enabled: cpu,logical_disk,net,os,service,system
      collector:
        service:
          include: windows_exporter
        scheduled_task:
          include: /Microsoft/.+
      log:
        level: debug
      scrape:
        timeout-margin: 0.5
      telemetry:
        path: /metrics
      web:
        listen-address: :9182
      ---
      #Note this is not an exhaustive list of all configuration values
      collectors:
        enabled: cpu,logical_disk,net,os,service,system
      collector:
        service:
          include: windows_exporter
        scheduled_task:
          include: /Microsoft/.+
      log:
        level: debug
      scrape:
        timeout-margin: 0.5
      telemetry:
        path: /metrics
      web:
        listen-address: :9182
      Code block. Example of a Windows Exporter configuration file
      Refer to the following to register a service using a configuration file.
      Color mode
      $ sc.exe create windows_exporter binPath= "C:\Temp\windows_exporter-0.31.3-amd64.exe --config.file=C:\Temp\config.yml" DisplayName= "Prometheus Windows Exporter" start= auto
      $ Start-Service windows_exporter
      $ sc.exe create windows_exporter binPath= "C:\Temp\windows_exporter-0.31.3-amd64.exe --config.file=C:\Temp\config.yml" DisplayName= "Prometheus Windows Exporter" start= auto
      $ Start-Service windows_exporter
      code block. Service registration using Windows Exporter configuration file
Reference
When using a configuration file together with command-line options, the values specified in the command-line options take precedence. Consequently, command-line options override the configuration file settings.
notice
If you have completed the Node Exporter configuration, you must install the OpenTelemetry Collector provided by ServiceWatch to finish the ServiceWatch Agent setup.
For detailed information, refer to ServiceWatch > ServiceWatch Agent 사용하기.

#Node Exporter metrics

Below is the collector and metric information that can be viewed through Node Exporter. It can be set as a Collector, or you can enable only specific metrics.

CategoryCollectorMetricDescription
Memorymeminfonode_memory_MemTotal_bytesTotal memory
Memorymeminfonode_memory_MemAvailable_bytesAvailable memory (used to determine if memory is insufficient)
Memorymeminfonode_memory_MemFree_bytesFree memory (empty memory)
Memorymeminfonode_memory_Buffers_bytesIO buffer
Memorymeminfonode_memory_Cached_bytesPage cache
Memorymeminfonode_memory_SwapTotal_bytesTotal swap
Memorymeminfonode_memory_SwapFree_bytesRemaining swap
Filesystemfilesystemnode_filesystem_size_bytesTotal file system size
Filesystemfilesystemnode_filesystem_free_bytestotal free space
Filesystemfilesystemnode_filesystem_avail_bytesSpace actually available to the user (available space for unprivileged users)
Table. Node Exporter key metrics

Node Exporter enables most collectors by default, but you can enable or disable only the collectors you want.

  • 메모리, 파일 시스템 Collector만 사용하고 싶을 때:
    Color mode
    ./node_exporter
      --collector.meminfo
      --collector.filesystem
    ./node_exporter
      --collector.meminfo
      --collector.filesystem
    code block. Enable specific collector in Node Exporter
  • Default 설정은 모두 해제하고 메모리, 파일 시스템 Collector만 사용하고 싶을 때:
    Color mode
    ./node_exporter
      --collector.disable-defaults
      --collector.meminfo
      --collector.filesystem
    ./node_exporter
      --collector.disable-defaults
      --collector.meminfo
      --collector.filesystem
    code block. Enable specific collector of Node Exporter
    Enable file system Collector for a specific mount point
    Color mode
    ./node_exporter
      --collector.disable-defaults
      --collector.filesystem.mount-points-include="/|/data"
    ./node_exporter
      --collector.disable-defaults
      --collector.filesystem.mount-points-include="/|/data"
    code block. Enable specific collector in Node Exporter
    Enable the file system Collector, excluding specific mount points
    Color mode
    ./node_exporter
      --collector.disable-defaults
      --collector.filesystem.mount-points-exclude="/boot|/var/log"
    ./node_exporter
      --collector.disable-defaults
      --collector.filesystem.mount-points-exclude="/boot|/var/log"
    code block. Enable specific collector in Node Exporter

파일 시스템 collector를 사용하고 싶지 않을때:

Color mode
./node_exporter --no-collector.filesystem
./node_exporter --no-collector.filesystem
code block. Disable specific collector in Node Exporter

Color mode
[Unit]
Description=Node Exporter
After=network-online.target

[Service]
User=nodeexp
ExecStart=/usr/local/bin/node_exporter
  --collector.disable-defaults
  --collector.meminfo
  --collector.filesystem

[Install]
WantedBy=multi-user.target
[Unit]
Description=Node Exporter
After=network-online.target

[Service]
User=nodeexp
ExecStart=/usr/local/bin/node_exporter
  --collector.disable-defaults
  --collector.meminfo
  --collector.filesystem

[Install]
WantedBy=multi-user.target
Code block. Node Exporter > /etc/systemd/system/node_exporter.service configuration

You can configure the Open Telemetry Collector to select and collect only the necessary metrics gathered from the Node Exporter. When you want to collect only specific metrics among those provided by a particular collector of Node Exporter, ServiceWatch를 위한 Open Telemetry Collector 사전 설정

Caution
Collecting metrics through the ServiceWatch Agent is classified as custom metrics, and unlike the default metrics collected from each service, it incurs charges; therefore, we recommend configuring only the metrics that are essential.

#Windows Exporter metrics

Below is the collector and metric information that can be viewed through the Windows Exporter. You can configure it as a Collector, or enable only specific metrics.

CategoryCollectorIndicator nameExplanation
Memorymemorywindows_memory_available_bytesAvailable memory
Memorymemorywindows_memory_cache_bytesCache memory
Memorymemorywindows_memory_committed_bytesCommitted memory
Memorymemorywindows_memory_commit_limitCommit limit
Memorymemorywindows_memory_pool_paged_bytespaged pool
Memorymemorywindows_memory_pool_nonpaged_bytesnon-paged pool
Disk informationlogical_diskwindows_logical_disk_free_bytesRemaining capacity
Disk informationlogical_diskwindows_logical_disk_size_bytesTotal capacity
Disk informationlogical_diskwindows_logical_disk_read_bytes_totalNumber of bytes read
Disk informationlogical_diskwindows_logical_disk_write_bytes_totalWrite byte count
Disk informationlogical_diskwindows_logical_disk_read_seconds_totalread latency
Disk informationlogical_diskwindows_logical_disk_write_seconds_totalwrite latency
Disk informationlogical_diskwindows_logical_disk_idle_seconds_totalidle time
Table. Windows Exporter Key Metrics

Windows Exporter enables most collectors by default, but you can configure only the collectors you want.

CPU, 메모리, 논리 디스크만 사용하고 싶을 때:

Color mode
#--collector.enabled option disables the default and activates only the specified Collector
.\windows_exporter.exe --collectors.enabled="memory,logical_disk"
#--collector.enabled option disables the default and activates only the specified Collector
.\windows_exporter.exe --collectors.enabled="memory,logical_disk"
code block. Enable a specific collector for Windows Exporter
Reference
Even without disabling unused collectors, using –collector.enabled will collect only the collectors specified in that option.

Color mode
#Register windows_exporter as a service
sc.exe create windows_exporter binPath= "C:\Temp\windows_exporter-0.31.3-amd64.exe --config.file=C:\Temp\config.yml" DisplayName= "Prometheus Windows Exporter" start= auto
#Service start
Start-Service windows_exporter
#Register windows_exporter as a service
sc.exe create windows_exporter binPath= "C:\Temp\windows_exporter-0.31.3-amd64.exe --config.file=C:\Temp\config.yml" DisplayName= "Prometheus Windows Exporter" start= auto
#Service start
Start-Service windows_exporter
code block. service registration
Color mode
#Note this is not an exhaustive list of all configuration values
collectors:
  enabled: logical_disk,memory
collector:
  service:
    include: windows_exporter
  scheduled_task:
    include: /Microsoft/.+
log:
  level: debug
scrape:
  timeout-margin: 0.5
telemetry:
  path: /metrics
web:
  listen-address: ":9182"
#Note this is not an exhaustive list of all configuration values
collectors:
  enabled: logical_disk,memory
collector:
  service:
    include: windows_exporter
  scheduled_task:
    include: /Microsoft/.+
log:
  level: debug
scrape:
  timeout-margin: 0.5
telemetry:
  path: /metrics
web:
  listen-address: ":9182"
Code block. Service configuration file

Through the Open Telemetry Collector configuration, you can set it to collect only the required metrics gathered from the Windows Exporter. When you want to collect only specific metrics among those provided by a particular collector of the Windows Exporter, [ServiceWatch를 위한 Open Telemetry Collector 사전 설정](/userguide/management/service_watch/how_to_guides/service_watch_agent/

Caution
Metric collection through the ServiceWatch Agent is classified as custom metrics and incurs charges, unlike the default metrics collected from each service; therefore, we recommend configuring only the metrics that are essential.

1.3 - API Reference

API Reference

1.4 - CLI Reference

CLI Reference

1.5 - Release Note

Virtual Server

2026.07.16
FEATURE Add server type and other feature changes
  • Even if the Backup service is connected, you can delete the server.
    • When a server is deleted, the schedule of the associated backup is removed, but the backup itself remains intact and can be restored and utilized.
  • s3 and h3 server types have been added.
    • Samsung Cloud Platform Sovereign will be offered after November 2026.

2026.05.21
CHANGED Shared Image generation size limit and AIOS integration suspension
  • Creating a shared Image is limited based on the OS disk size.
    • If the OS disk size is 500 GB or larger, you cannot create a shared Image.
  • The AIOS service integration has been stopped.
    • Since the AIOS service has been discontinued, the AIOS integrated service (LLM Endpoint) has been removed from Virtual Server.
2026.03.19
FEATURE Add standard image, add SSD_Provisioned disk type, and add ServiceWatch metric monitoring capability
  • Additional OS Image provision
    • Standard Image has been added. * (Window server 2016)
  • An SSD volume with configurable IOPS and Throughput has been added.
    • When creating Block Storage, you can select the SSD_Provisioned disk type.
    • You can set the maximum IOPS and Throughput.
  • You can view the Virtual Server ServiceWatch metric monitoring graph on the detail page.
2025.12.16
FEATURE Virtual Server feature addition
  • Additional OS Image provision
    • Standard Image has been added. * (Alma Linux 9.6, Oracle Linux 9.6, RHEL 9.6, Rocky Linux 9.6)
  • Add new Server Group policy
    • Partition(Virtual Server and Block Storage distributed placement) policy has been added.
  • You can install the Virtual Server ServiceWatch Agent to collect custom metrics and logs.
2025.10.23
FEATURE Add server name change feature and provide ServiceWatch service integration
  • You can change the server name on the Virtual Server detail page in the Samsung Cloud Platform Console.
    • When the server name is changed, the OS hostname remains unchanged, and only the information within the Samsung Cloud Platform Console is updated.
  • ServiceWatch service integration provision
    • You can monitor data through the ServiceWatch service.
2025.07.01
FEATURE Add Virtual Server feature and change Image sharing method
  • Add Virtual Server feature
    • IP, Public NAT IP, Private NAT IP configuration feature has been added.
    • An LLM Endpoint is provided for LLM usage.
    • When creating a Virtual Server, you can select an OS Image subscribed from the Marketplace.
    • The second-generation server type has been added.
      • Add a 2nd-generation (s2) server type based on the Intel 4th-generation (Sapphire Rapids) Processor. * For detailed information, see Virtual Server 서버 타입.
  • The method for sharing images between accounts has changed.
    • qcow2 Image or shared Image can be newly created and shared.
2025.02.27
FEATURE NAT configuration feature and addition of OS Image, server type
  • Add Virtual Server feature
    • The NAT configuration feature has been added to Virtual Server.
    • Additional OS Image provision
      • Standard Image has been added. * (Alma Linux 8.10, Alma Linux 9.4, Oracle Linux 8.10, Oracle Linux 9.4, RHEL 8.10, RHEL 9.4, Rocky Linux 8.10, Rocky Linux 9.4, Ubuntu 24.04)
      • An image for Kubernetes has been added. * You can create a Kubernetes Engine using an image for Kubernetes.
    • Add second-generation server type
      • Add 2nd-generation (h2) server type based on Intel 4th-generation (Sapphire Rapids) Processor. * For detailed information, see Virtual Server 서버 타입.
  • Samsung Cloud Platform Common Feature Change
    • Account, IAM, and Service Home, tags, etc., have been updated to reflect common CX changes.
2024.10.01
NEW Virtual Server service official version release
  • Virtual Server service has been officially launched.
  • We have launched a virtualized server that can be freely allocated and used as needed at the required time without having to purchase infrastructure resources individually.
2024.07.02
NEW Beta version release
  • We have launched a virtualized server that can be freely allocated and used as needed at the required time without having to purchase infrastructure resources individually.
2025.10.23
FEATURE Add server name change feature and provide ServiceWatch service integration
  • You can change the server name on the Virtual Server detail page in the Samsung Cloud Platform Console.
    • When the server name is changed, the OS hostname remains unchanged, and only the information within the Samsung Cloud Platform Console is updated.
  • ServiceWatch service integration provision
    • You can monitor data through the ServiceWatch service.
2025.07.01
FEATURE Add Virtual Server feature and change Image sharing method
  • Add Virtual Server feature
    • IP, Public NAT IP, Private NAT IP configuration feature has been added.
    • An LLM Endpoint is provided for LLM usage.
    • When creating a Virtual Server, you can select an OS Image subscribed from the Marketplace.
    • The second-generation server type has been added.
      • Add a 2nd‑generation (s2) server type based on the Intel 4th‑generation (Sapphire Rapids) Processor. * For detailed information, see Virtual Server 서버 타입.
  • The method for sharing images between accounts has changed.
    • qcow2 Image or shared Image can be newly created and shared.
2025.02.27
FEATURE NAT configuration feature and addition of OS Image, server type
  • Add Virtual Server feature
    • The NAT configuration feature has been added to Virtual Server.
    • Additional OS Image provision
      • Standard Image has been added. * (Alma Linux 8.10, Alma Linux 9.4, Oracle Linux 8.10, Oracle Linux 9.4, RHEL 8.10, RHEL 9.4, Rocky Linux 8.10, Rocky Linux 9.4, Ubuntu 24.04)
      • An image for Kubernetes has been added. * You can create a Kubernetes Engine using an image for Kubernetes.
    • Add second-generation server type
      • Add 2nd-generation (h2) server type based on Intel 4th-generation (Sapphire Rapids) Processor. * For detailed information, see Virtual Server 서버 타입.
  • Samsung Cloud Platform Common Feature Change
    • Account, IAM, and Service Home, tags, etc., have been updated to reflect common CX changes.
2024.10.01
NEW Virtual Server service official version release
  • Virtual Server service has been officially launched.
  • We have launched a virtualized server that can be freely allocated and used as needed at the required time without having to purchase infrastructure resources individually.
2024.07.02
NEW Beta version release
  • We have launched a virtualized server that can be freely allocated and used as needed at the required time without having to purchase infrastructure resources individually.

2 - Virtual Server Auto-Scaling

2.1 - Overview

Service Overview

Virtual Server Auto-Scaling is a service that automatically scales resources up or down based on demand. You can add or terminate servers running the application according to predefined conditions or schedules.
An Auto-Scaling Group uses a pre-created Launch Configuration as a configuration template to launch servers, and can adjust and manage the number of servers. It adjusts to ensure the number of servers does not fall below the specified minimum or exceed the specified maximum.
If you register a schedule with an Auto-Scaling Group, you can set the number of servers according to the specified schedule. If you register a policy, you can increase or decrease the number of servers based on predefined conditions.

Features

  • Easy and convenient computing environment setup: Through a web-based Console, users can easily configure the required computing environment themselves via Self Service, from creating Launch Configurations to creating/modifying/deleting Auto-Scaling Groups.

  • Elastic Resource Usage: Computing resources can be used elastically according to the service load and usage. Users can schedule resource usage for predictable specific time periods, and can adjust resource consumption to accommodate temporary access by an unspecified large number of users.

  • Availability Improvement: Virtual Server Auto-Scaling provides a function that adjusts resources to match variable demand so that the traffic required by the user can always be handled. Through this, users can achieve improved application performance and availability.

  • Maximize Cost Savings: You can reduce unnecessary expenses by using resources only as needed according to demand fluctuations. By flexibly allocating resources in response to traffic increases or decreases during specific periods such as nights, weekends, and month-ends, you can maximize cost-saving effects.

Service Architecture Diagram

Diagram
Figure. Virtual Server Auto-Scaling Diagram

Provided features

Virtual Server Auto-Scaling provides the following features.

  • Launch Configuration: It is a configuration template used to create a Virtual Server in an Auto-Scaling Group. When creating a Launch Configuration, you set information about the Virtual Server such as the image, server type, Key Pair, Block Storage, etc.
  • Server Count Adjustment: Provides several ways to adjust the number of servers. By using policies, you can add a Virtual Server when load exceeds a threshold and release the Virtual Server when demand is low, maintaining application availability while reducing costs. You can also add and release Virtual Servers according to a predefined schedule, and manually adjust the number of servers in an Auto-Scaling Group as needed.
  • Load Balancer Integration: You can use a Load Balancer to evenly distribute application traffic across Virtual Servers. Whenever a Virtual Server is added or removed, it is automatically registered with or deregistered from the Load Balancer.
  • Network Connection: You can connect the Auto-Scaling Group’s standard subnet, automatic IP allocation, and Public NAT IP. Provides a local subnet connection for inter-server communication.
  • Security Group Application: Security Group is a virtual logical firewall that controls inbound/outbound traffic generated on a Virtual Server. Inbound rules control incoming traffic to the Virtual Server, and outbound rules control outgoing traffic from the Virtual Server.
  • Monitoring: You can view monitoring data such as CPU, Memory, and Disk of Virtual Servers created in an Auto-Scaling Group via the ServiceWatch service. Based on the monitoring data, you can set load thresholds using Auto-Scaling policies, and when thresholds are exceeded, you can add or remove servers.
Reference
The policy of the Auto-Scaling Group will be integrated with ServiceWatch starting March 25, 2026. The policies of an Auto-Scaling Group are created in conjunction with ServiceWatch alarm policies, and alarm policies linked to an Auto-Scaling Group cannot be modified or deleted in ServiceWatch; they can only be managed through the Auto-Scaling Group.

Component

Virtual Server Auto-Scaling creates an Auto-Scaling Group through a Launch Configuration and monitors and manages the servers.

Launch Configuration

This is a Configuration template used to create a Virtual Server in an Auto-Scaling Group. The main features are as follows.

  • Image: Provides OS standard images and Custom images created by the user. Users can select and use them according to the service they wish to configure.
  • Keypair: Provides the Keypair method for secure OS access.
  • Init script: Users can define a script to be executed when the Virtual Server starts.
  • For more details, refer to Launch Configuration Create.
Reference
In Launch Configuration, please refer to the Virtual Server OS Image Provisioning Versions and Virtual Server Server Types for the selectable images and server types.

Auto-Scaling Group

Launch Configuration is used as a pre-configuration template for creating servers. By creating an Auto-Scaling Group, you can adjust and manage the number of servers. The main features are as follows.

  • Launch Configuration: A Configuration template used to create a Virtual Server in an Auto-Scaling Group.
  • Server Count Setting: Virtual Server Auto-Scaling provides several ways to adjust the number of servers in an Auto-Scaling Group.
    • Fixed Server Count Method: When creating an Auto-Scaling Group, this method keeps the default settings using the configured number of servers without any additional schedules or policies. Refer to Create Auto-Scaling Group to set the Min, Desired, and Max server counts.
    • Server Count Manual Adjustment Method: In an Auto-Scaling Group, you can increase or decrease the number of servers by modifying the server count to the desired amount. You can choose whether to manually set the desired server count. Please refer to 서버 수 수정하기.
    • Schedule reservation method: You can schedule daily, weekly, monthly, or one-time, and set the desired number of servers at the specified time. This is useful when you can predict when to scale the number of servers up or down. If you use the schedule method, please refer to Manage Schedules to add and manage schedules.
    • Policy Mode: You can use a policy to dynamically adjust servers. When a monitoring metric exceeds a configured threshold, the number of servers is adjusted. At this time, you can choose one of three methods to adjust the server count: increase or decrease by a specified number, increase or decrease by a specified percentage, or fix the server count to a given value. When servers are started or terminated due to a policy, the monitoring metric (CPU utilization) may temporarily exceed the threshold registered in the policy. However, because this is a brief moment, a cooldown period is set to avoid treating it as an abnormal situation. If you want to use the policy mode, see Managing Policies.
  • Load Balancer: Automatically connects and disconnects to the Load Balancer registered in the Auto-Scaling Group whenever a Virtual Server is added or terminated.
Reference
The Load Balancer of the Auo-Scaling Group will operate in conjunction with the Load Balancer starting in February 2025.

Constraints

The constraints of Virtual Server Auto-Scaling are as follows.

CategoryExplanation
Number of Virtual Servers per Auto-Scaling Group50 or fewer
Number of policies per Auto-Scaling Group12 or fewer
Number of schedules per Auto-Scaling Group20 or fewer
Number of LB server groups and ports per Auto-Scaling Group3 or fewer
Table. Virtual Server Auto-Scaling Group Constraints
Caution
  • If the Image you are using is a discontinued standard Image, Scale-out will not work.
    If the Image you are using is Custom Image, Scale out will continue to operate correctly even after that version is discontinued.
  • We recommend replacing the Launch Configuration with the latest version of the Image or a Custom Image before the current Image reaches end of support.
  • For detailed information about the OS Image provided by Virtual Server, see OS Image Provided Versions.

Prior Service

This is a list of services that must be pre-configured before creating the service. Please refer to the guide provided for each service and prepare in advance.

Service CategoryserviceDetailed description
NetworkingVPCA service that provides an isolated virtual network in a cloud environment
NetworkingSecurity GroupVirtual firewall that controls server traffic
Table. Virtual Server Auto-Scaling Prerequisite Service

2.1.1 - Monitoring Metrics

information
The Auto-Scaling Group policy will be linked with the ServiceWatch alarm policy starting March 25, 2026. Please refer to Virtual Server Auto-Scaling > ServiceWatch Metrics.
Cloud Monitoring service termination notice

According to Samsung Cloud Platform’s policy, the Cloud Monitoring service is scheduled to be discontinued in September 2026.
Accordingly, after the September 2026 release, resource monitoring of the Samsung Cloud Platform via Cloud Monitoring will no longer be possible.

With the new alternative service, you can continuously perform resource monitoring by using ServiceWatch, released in October 2025.
ServiceWatch provides more modern and powerful features, replacing Cloud Monitoring to deliver a smooth monitoring environment.

Detailed information about ServiceWatch is available in the ServiceWatch Overview.

Virtual Server Auto-Scaling is a service provided for Virtual Servers that offers individual Virtual Server monitoring metrics and monitoring metrics provided by policies based on Cloud Monitoring.

Virtual Server Monitoring Metrics

The table below shows the monitoring metrics of Virtual Server that can be viewed through Cloud Monitoring. For detailed usage of Cloud Monitoring, refer to the Cloud Monitoring guide.

Memory-related metrics are not provided for Windows OS.

Performance itemsDetailed descriptionunit
Memory Total [Basic]bytes of usable memorybytes
Memory Used [Basic]bytes of currently used memorybytes
Memory Swap In [Basic]bytes of the replaced memorybytes
Memory Swap Out [Basic]bytes of the replaced memorybytes
Memory Free [Basic]bytes of unused memorybytes
Disk Read Bytes [Basic]Read bytesbytes
Disk Read Requests [Basic]Number of read requestscnt
Disk Write Bytes [Basic]write bytesbytes
Disk Write Requests [Basic]Number of write requestscnt
CPU Usage [Basic]Average system CPU usage over 1 minute%
Instance State [Basic]Instance statusstate
Network In Bytes [Basic]Received bytesbytes
Network In Dropped [Basic]Incoming packet dropcnt
Network In Packets [Basic]Number of received packetscnt
Network Out Bytes [Basic]sent bytesbytes
Network Out Dropped [Basic]Transmit packet dropcnt
Network Out Packets [Basic]Number of transmitted packetscnt
Table. Virtual Server Monitoring Metrics (Provided by default)

Monitoring metrics provided by Cloud Monitoring-based policies

The table below shows the monitoring metrics provided by the policy of a Cloud Monitoring‑based Auto‑Scaling Group. For detailed information on policy settings based on Cloud Monitoring, see 정책 관리하기.

Performance itemsDetailed descriptionunit
CPU Usage [Basic]Average system CPU usage over 1 minute%
Memory Used [Basic]bytes of currently used memorybytes
Network In Bytes [Basic]Received bytesbytes
Network In Packets [Basic]Number of received packetscnt
Network Out Bytes [Basic]sent bytesbytes
Network Out Packets [Basic]Number of transmitted packetscnt
Table. Monitoring metrics provided by Cloud Monitoring-based policies

2.1.2 - ServiceWatch Metrics

Virtual Server Auto-Scaling is a service provided for Virtual Servers that offers individual Virtual Server monitoring metrics and monitoring metrics supplied by ServiceWatch-based policies.

Virtual Server Monitoring Metrics

The basic metrics provided by Virtual Server can be found in ServiceWatch > Virtual Server Basic Metrics.

Reference
For checking metrics in ServiceWatch, refer to the ServiceWatch guide.

ServiceWatch monitoring metrics provided by the Auto-Scaling Group policy

The table below shows the ServiceWatch monitoring metrics provided by the Auto-Scaling Group policy. For detailed information on configuring Auto-Scaling Group policies, see Managing Policies.

Performance itemsDetailed descriptionunit
CPU UsageCPU usagePercent
Network In BytesReceived bytes on the network interfaceBytes
Network In PacketsNumber of packets received on the network interfaceCount
Network Out BytesData transmitted on the network interface (bytes)Bytes
Network Out PacketsNumber of packets transmitted on the network interfaceCount
Table. ServiceWatch monitoring metrics provided by the Auto-Scaling Group policy

2.2 - How-to guides

Users can create the service by entering the required information for the Auto-Scaling Group and selecting detailed options through the Samsung Cloud Platform Console.

Create Auto-Scaling Group

You can create and use the Auto-Scaling Group service from the Samsung Cloud Platform Console.

Reference
To create an Auto-Scaling Group, you need to create a Launch Configuration in advance. Refer to Launch Configuration 생성하기.

To create an Auto-Scaling Group, follow the steps below.

  1. Click the All Services > Compute > Virtual Server menu. 1. Navigate to the Service Home page of the Virtual Server.

  2. Click the Auto-Scaling Group menu. 2. Auto-Scaling Group List Navigate to the page.

  3. On the Auto-Scaling Group List page, click the Create Auto-Scaling Group button. 3. Auto-Scaling Group Creation Navigate to the page.

  4. On the Auto-Scaling Group creation page, enter the information required to create the service.

    • In the Launch Configuration area, select the Launch Configuration.
      • Click the Launch Configuration creation button to create a new Launch Configuration.
    • Enter or select the required information in the Service Information Input area.
      Category
      required status
      Detailed description
      Auto-Scaling Group nameRequiredAuto-Scaling Group name
      • manage servers of the same type and purpose as a group
      Server nameRequiredServer name to be created within the Auto-Scaling Group
      • Automatically assign a server name by combining the entered server name, which serves as an identifier to distinguish servers created within the Auto-Scaling Group, with a sequence
      Number of serversRequiredNumber of servers to create in the Auto-Scaling Group
      • Enter a value between 0 and 20 (Min≤Desired≤Max)
      • Min: Set the minimum number of servers the Auto-Scailg Group should maintain
      • Desired: Set the target number of servers within the Auto-Scailg Group; also indicates the number of servers initially created when the Auto-Scailg Group is created
      • Max: Set the maximum number of servers the Auto-Scailg Group can maintain
      • After creating the Auto-Scaling Group, you can adjust the settings using the Edit button. For details, see 서버 수 수정하기
      Desired server count manual settingSelectionSelect whether to manually change the Desired server count
      Network Settings > Network SettingsRequiredNetwork settings for Auto-Scaling Group
      • Select the desired VPC and standard Subnet
      • IP can only be auto-generated.
      • When you select local Subnet, you can choose the desired local Subnet, and IP can only be auto-generated
      Network Settings > Security GroupSelectionTo allow required connections, you need to configure a Security Group
      • If you do not configure a Security Group, all inbound and outbound traffic will be blocked according to the default rule (Any/Deny)
      • For Linux servers, allow SSH traffic
      • For Windows servers, allow RDP traffic
      Load BalancerSelectionConnect the Auto-Scaling Group to a Load Balancer
      • Register the Auto-Scaling Group’s servers as members of the LB server group
      • LB server group: Select the LB server group that exists in the chosen VPC
      • Port: Enter a value between 1 and 65,534
      • Press the + button to add an LB server group (up to a total of three LB server groups and ports are allowed)
      • LB server groups that use Weighted Round Robin or Weighted Least Connection load balancing cannot be selected
      • Draining Timeout value: After checking Draining Timeout as Enabled, you can set the Draining Timeout value
        • Draining Timeout: The waiting period before removing a server from the Load Balancer
          • Since sessions may remain connected to the server, setting a Draining Timeout and waiting allows safe session cleanup
        • If the Load Balancer is unused, Draining Timeout cannot be configured
        • The default is 300 seconds; you can enter values from a minimum of 1 second up to a maximum of 3,600 seconds.
      Table. Auto-Scaling Group service information entry items
    • In the Scaling policy configuration area, set the Scaling policy.
      • For detailed information on policy settings, see 정책 추가하기.
        Category
        required status
        Detailed description
        Current settingSelectionSet the Scaling policy now
        • Add Policy button click displays policy information input fields
        Set laterSelectionAfter creating an Auto-Scaling Group, set the policy on the details page
        Table. Auto-Scaling Group Scaling policy configuration items
    • In the Notification Settings area, configure the notification recipients and notification method.
      • For detailed information about notification settings, refer to 알림 추가하기.
        Category
        required status
        Detailed description
        Current settingSelectionSet the notification recipients and notification method now
        • Click the Add Notification button to open the Add Notification popup
        • For detailed information on notification settings, refer to the details
        • You can modify notification information by clicking the Edit button in the notification recipients list
        Set laterSelectionAfter creating an Auto-Scaling Group, set the notification recipients and notification method on the details page.
        Table. Auto-Scaling Group notification configuration items
    • In the Additional Information Input area, enter or select the required information.
      Category
      required status
      Detailed description
      Status checkSelectionCheck the status of Virtual Server and Load Balancer and replace servers that are in an Unhealthy state
      • Virtual Server status check: Enabled is the default and cannot be changed
      • Load Balancer status check: Enabled only when a Load Balancer is connected in the service information input area
      • Grace period: Set the time to defer status checks until the newly added server operates normally
        • Enter 0 to disable the status check grace period
      tagSelectionAdd Tag
      • Up to 50 per resource can be added
      • Add Tag button after clicking, enter or select Key, Value values
      Table. Auto-Scaling Group additional information input fields
      Caution
      A server that the user changes to Stop is not considered Unhealthy, so even when using the health check feature, the server will not be replaced.
  5. Summary Check the detailed information and estimated billing amount generated in the panel, and click the Create button.

    • When creation is complete, verify the created Auto-Scaling Group on the Auto-Scaling Group list page.

Auto-Scaling Group Check detailed information

The Auto-Scaling Group service allows you to view and edit the full resource list and detailed information. Auto-Scaling Group Details page is organized into Detailed Information, Policies, Schedule, Virtual Server, Load Balancer, Tags, Operation History tabs.

To view detailed information about the Auto-Scaling Group, follow these steps.

  1. All Services > Compute > Virtual Server menu, click. 1. Navigate to the Service Home page of the Virtual Server.
  2. Click the Auto-Scaling Group menu. 2. Navigate to the Auto-Scaling Group List page.
  3. Auto-Scaling Group List page, click the resource to view detailed information. 3. Go to the Auto-Scaling Group Details page.
    • Auto-Scaling Group Details The page displays status information and additional feature information, and Details, Policies, Schedule, Virtual Server, Load Balancer, Tags, Operation History are organized as tabs.
      CategoryDetailed description
      Auto-Scaling Group statusStatus of the Auto-Scaling Group created by the user
      • Creating: Creating Auto-Scaling Group
      • In Service: In Service
      • Scale In: Scale In in progress
      • Scale Out: Scale Out in progress
      • Cool Down: Cooling down
      • Terminating: Terminating Auto-Scaling Group
      • Attach to LB: Attaching to Load Balancer
      • Detach from LB: Detaching from Load Balancer
      Delete Auto-Scaling GroupButton to delete Auto-Scaling Group
      Table. Auto-Scaling Group status information and additional features

Detailed Information

Auto-Scaling Group list page allows you to view detailed information of the selected resource and, if needed, modify the information.

CategoryDetailed description
serviceservice name
Resource TypeResource Type
SRNUnique resource ID in Samsung Cloud Platform
  • In Auto-Scaling Group, it means the Auto-Scaling Group SRN
Resource nameResource name
  • In an Auto-Scaling Group, it refers to the Auto-Scaling Group name
Resource IDUnique resource ID in the service
ConstructorUser who created the service
Creation date and timeService creation date and time
ModifierUser who edited the service information
Modification date and timeDate and time the service information was modified
Auto-Scaling Group nameAuto-Scaling Group name
Launch Configuration nameWhen creating an Auto-Scaling Group, the selected Launch Configuration name
  • the Image template used when creating the Virtual Server of the Auto-Scaling Group
Number of serversCurrent number of servers created in the Auto-Scaling Group and the configured Min, Desired, Max server counts
Desired server count manual settingDesired server count manual setting enable/disable
VPCAuto-Scaling Group VPC information
Standard SubnetGeneral Subnet and NAT IP usage information for Auto-Scaling Group
Local SubnetLocal Subnet of Auto-Scaling Group
Security GroupAuto-Scaling Group’s Security Group
Status checkWhether to use the replacement function for Virtual Server and Load Balancer in an Unhealthy state
  • Virtual Server: Use is the default and cannot be changed
  • Load Balancer: Indicates usage when a Load Balancer is attached
    • Use*: When a Load Balancer is connected and health check is used
    • Not used: When a Load Balancer is connected and health check is not used
    • Use (does not operate because no LB is currently connected): When a Load Balancer is connected and health check is in use, but the Load Balancer has been detached
    • -: When no Load Balancer is connected
  • Grace period: Time to defer health checks
    • If 0, the health check grace period is not used
  • Click the Edit button to configure whether Load Balancer health check is used and set the Grace period. For details, see Change health check
Table. Auto-Scaling Group Details – Details Tab Items
Caution
A server that the user changes to Stop is not considered Unhealthy, so even when using the health check feature, the server will not be replaced.

Policy

On the Auto-Scaling Group List page, you can view the list of policies for the selected resource and, if needed, add or manage policies.

CategoryDetailed description
CategoryPolicy Type
  • Scale In: Return server count
  • Scale Out: Increase server count
Policy NameName for policy classification
Execution conditionsConditions to trigger the policy
  • Statistic: Method to calculate the Metric Type
    • Average: Average of servers in the Auto-Scaling Group
    • Min: Minimum value among servers in the Auto-Scaling Group
    • Max: Maximum value among servers in the Auto-Scaling Group
  • Metric Type: CPU Usage, Network In(Bytes), Network Out(Bytes), Network In(Packets), Network Out(Packets)
  • Operator: >= > <= <
  • Threshold: Threshold corresponding to the Metric Type
  • Period: Continuous duration required to trigger the condition (the condition must be continuously met for N minutes for the policy to execute)
execution unitPolicy execution method
  • Policy Type: Select the type of policy to execute
    • Increase or decrease the number of servers by a specified count: Increase or reduce the server count to the Target Value
    • Increase or decrease the number of servers by a specified ratio: Increase or reduce by the Target Value ratio
    • Fix the number of servers to the entered value: Fix the server count according to the Target Value
  • Target Value: The number or ratio to execute the selected Policy Type
CooldownThe waiting time (seconds) when a server is started or terminated due to a policy
  • The default is 300 seconds, configurable from a minimum of 1 second up to a maximum of 3,600 seconds
More > EditModify the policy information
More > ActivateEnable this policy
  • Can be selected only when the policy is disabled
More > DisableDisable this policy
  • Can be selected only when the policy is active
Table. Auto-Scaling Group Details - Policy Tab Items
Reference
For policy management and policy example explanations, please refer to 정책 관리하기.

Schedule

Auto-Scaling Group List page allows you to view the schedule list of the selected resources and, if necessary, add or manage schedules.

CategoryDetailed description
nameSchedule name
MinMinimum number of servers set in the schedule
DesiredNumber of target servers set in the schedule
MaxMaximum number of servers set in the schedule
PeriodSchedule execution frequency
  • set to one of daily/weekly/monthly/once
Date/DaySchedule execution date or day of week
  • Date/Day selected according to the schedule cycle
Execution timeSchedule execution time
Time zoneSchedule execution time window
statusSchedule status
More > EditEdit the schedule information
More > ActivateActivate this schedule
  • Can be selected only when the schedule is disabled
More > DisableDisable this schedule
  • Can be selected only when the schedule is active
Add scheduleAdd a new schedule
DeleteDelete the selected schedule from the list
Table. Auto-Scaling Group Details - Schedule Tab Items

Reference
For detailed information on schedule management, refer to 스케줄 관리하기.

Virtual Server

On the Auto-Scaling Group List page, you can view the Virtual Server list of the selected resource.

CategoryDetailed description
Server nameServer name created in the Auto-Scaling Group
  • Server name when clicked, navigate to the Virtual Server Details page
IPIP assigned to the server
Creation date and timeThe date and time the server was created
statusResult of status check for Virtual Server and Load Balancer
  • Healthy: When both Virtual Server and Load Balancer are in Healthy state
  • Unhealthy: When one or more of Virtual Server or Load Balancer are in Unhealthy state
    • Unhealthy: Including cases where one or more of the three Load Balancer server groups connected to the Virtual Server are Unhealthy
    • The cause of Unhealthy can be found in the Operation History tab
Table. Auto-Scaling Group Details - Virtual Server Tab Items

Load Balancer

Auto-Scaling Group List page allows you to view the Load Balancer list of the selected resources, and, if necessary, you can add or manage Load Balancers.

CategoryDetailed description
Draining TimeoutWhether to use Draining Timeout
  • When you click the Edit button, set the Draining Timeout usage
  • If it is already in use, change the time
  • If the Load Balancer is not in use, Draining Timeout cannot be configured.
Load BalancerLoad Balancer usage
  • When you click the Edit button, set the Load Balancer usage
  • If it is already in use, add or modify the Load Balancer
  • Up to three LB server groups can be added
Load Balancer > Load Balancer nameLoad Balancer name to attach to the Auto-Scaling Group
Load Balancer > LB server groupLoad Balancer’s LB server group
  • LB server groups that use Weighted Round Robin or Weighted Least Connection load balancing cannot be selected
Load Balancer > PortPort registered as a Member in the LB server group
Table. Auto-Scaling Group Details – Load Balancer Tab Items
Reference

Notification

On the Auto-Scaling Group List page, you can view the notification recipients and notification methods for the selected resource.

CategoryDetailed description
Notification recipientName of the notification recipient
emailNotification recipient’s email
Create ServerWhether to send a notification when a server creation-related alert occurs
  • On success: Whether to send when creation succeeds
  • On failure: Send when creation fails
Server terminationWhether to send a notification when a server termination-related alert occurs
  • On success: Whether to send when termination succeeds
  • On failure: Send when termination fails
When the policy execution condition is metWhether a notification is triggered when the policy execution conditions are met
statusNotification activation status
  • Active: enabled
More > EditEdit the notification information
More > ActivateEnable this notification information
  • Can be selected only when the notification is disabled
More > DisableDisable the notification information
  • Can be selected only when the notification is enabled
Add notificationAdd a new notification
DeleteDelete the selected notification from the list
Table. Auto-Scaling Group Details - Notification Tab Items
Reference
For detailed information about notification settings, refer to 알림 관리하기.

Tag

On the Auto-Scaling Group List page, you can view the tag information of the selected resource, and you can add, modify, or delete it.

CategoryDetailed description
Tag listTag list
  • You can view the Key and Value information of the tag
  • Up to 50 tags can be added per resource
  • When entering a tag, you can search and select from the list of previously created Keys and Values
Table. Auto-Scaling Group Details - Tag Tab Items

Operation History

Auto-Scaling Group List page allows you to view the operation history of the selected resource.

CategoryDetailed description
Task History ListResource Change History
  • Operation Timestamp, Resource ID, Resource Name, Operation Details, Event Topic, Operation Result, Check Operator Information
Table. Auto-Scaling Group Details - Task History Tab Items

Auto-Scaling Group Managing Resources

If you need management functions for the created Auto-Scaling Group, you can perform tasks on the Auto-Scaling Group Details page.

Edit Launch Configuration

You can modify the Launch Configuration of an Auto-Scaling Group.

참고
Even if you modify the Launch Configuration, it does not apply to servers already created in the Auto-Scaling Group; it only takes effect for servers created thereafter. If you want to apply the updated Launch Configuation to all servers in the Auto-Scaling Group, set the server count (Desired) to 0 to delete all existing servers, then change the server count (Desired) to the desired amount.

To modify the Launch Configuration of an Auto-Scaling Group, follow these steps.

  1. All Services > Compute > Virtual Server Click the menu. 1. Navigate to the Service Home page of the Virtual Server.

  2. Click the Auto-Scaling Group menu. 2. Go to the Auto-Scaling Group List page.

  3. On the Auto-Scaling Group List page, click the resource to edit the Launch Configuration. 3. Auto-Scaling Group Details Navigate to the page.

  4. Click the Edit button for the Launch Configuration name. 4. Launch Configuration Edit The popup window opens. 4. You can view the list of selectable Launch Configurations.

    Category
    Detailed description
    Launch Configuration nameLaunch Configuration name
    ImageLaunch Configuration OS Image
    Server typeLaunch Configuration server type
    Block StorageLaunch Configuration Block Storage Settings
    Auto-Scaling Group countNumber of Auto-Scaling Groups with a Launch Configuration applied
    View DetailsButton to view Launch Configuration details
    Table. Launch Configuration list items

  5. Launch Configuration Edit In the popup window, select the Launch Configuration to change, then click the Confirm button. 5. Launch Configuration Modification Notification A popup window opens. 5. Launch Configuration Modification Notification Check the message in the popup window and click the Confirm button.

Modify server count

You can modify the number of servers in the Auto-Scaling Group.

Reference
The maximum number of servers that can be configured is 20. However, when a Load Balancer is present, exclude servers that are not connected to the Load Balancer.

To modify the number of servers in an Auto-Scaling Group, follow these steps.

  1. All Services > Compute > Virtual Server Click the menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. Click the Auto-Scaling Group menu. 2. Navigate to the Auto-Scaling Group List page.
  3. On the Auto-Scaling Group List page, click the resource to modify the server count. 3. Go to the Auto-Scaling Group Details page.
  4. Click the Edit server count button. 4. Edit Server Count popup window opens.
  5. Edit Server Count After entering the required fields in the popup window, click the Confirm button.
    Category
    required status
    Detailed description
    Number of servers > MinRequiredMinimum number of servers to modify
    • Set the number of servers that the Auto-Scaling Group maintains at a minimum
    Number of servers > DesiredRequiredNumber of target servers to modify
    • Set the number of target servers within the Auto-Scaling Group
    Number of servers > MaxRequiredTarget number of servers to modify
    • Set the number of servers that the Auto-Scailg Group can maintain at maximum
    Table. Auto-Scailg Group server count modification items

Terminate Virtual Server Created by Auto-Scaling Group

Virtual Server created in the Auto-Scaling Group can be terminated by clicking the Terminate button on its detail page or by modifying the server count.

Caution
  • Please note that data cannot be recovered after terminating the service.
  • To cancel the service, first disconnect the File Storage and disable the Lock. * If the File Storage is connected or a lock is set, the Virtual Server cannot be terminated.
Reference
  • If you terminate a Virtual Server created in the Auto-Scaling Group, the associated Load Balancer is automatically detached.
  • When a Virtual Server is terminated, the status of the attached Storage is as follows.
    • If Delete on termination is not set: the volume will not be deleted even if the Virtual Server is terminated.
    • When Delete on termination is set: If you terminate the Virtual Server, the volume will be deleted. * However, if a Snapshot exists, it will not be deleted even if Delete on termination is set.
    • Multi-attach volume: It can only be deleted when the server you are trying to delete is the last remaining server attached to the volume.

Cancel on the Virtual Server detail page

To cancel a Virtual Server on the Virtual Server detail page, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. On the Service Home page, click the Virtual Server menu. 2. Go to the Virtual Server List page.
  3. On the Virtual Server List page, select the Virtual Server created by the Auto-Scaling Group. 3. Navigate to the Virtual Server Details page.
  4. On the Virtual Server Details page, click the Service Cancellation button. 4. A popup window announcing the server termination opens.
  5. Confirm Click the button. 5. Server termination is complete.

Terminate by adjusting the server count

To terminate a Virtual Server by adjusting the server count, follow the steps below.

  1. All Services > Compute > Virtual Server Click the menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. Click the Auto-Scaling Group menu. 2. Go to the Auto-Scaling Group List page.
  3. Auto-Scaling Group List page, click the resource to terminate the Virtual Server. 3. Auto-Scaling Group Details page.
  4. Click the Edit button for the Server Count item. 4. Edit Server Count popup window opens.
  5. Server Count Modification In the popup window, after reducing the Desired count, click the Confirm button. 5. When the desired number of servers is adjusted, the Virtual Server is terminated.
Notice
  • Desired server count manual setting is unused, you cannot modify the Desired server count. * To modify the Desired server count, refer to Desired 서버 수 수동 설정 수정하기.
  • When a Virtual Server is terminated, the Desired server count remains unchanged, and scale-out proceeds according to the Desired check batch.

Manually modify the desired number of servers

You can manually change the Desired server count setting of the Auto-Scaling Group.

Reference
If you do not select Use for manual Desired server count setting, you cannot modify the Desired server count in the details’ server count edit. The desired number of servers is automatically adjusted by checking the Virtual Server status every 10 minutes.

To modify the manual setting of the Desired server count for an Auto-Scaling Group, follow these steps.

  1. All Services > Compute > Virtual Server Click the menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. Click the Auto-Scaling Group menu. 2. Navigate to the Auto-Scaling Group List page.
  3. Auto-Scaling Group List page, click the resource to manually change the Desired server count. 3. Auto-Scaling Group Details Navigate to the page.
  4. Click the Edit server count button. 4. Desired server count manual setting The popup window opens.
  5. Desired server count manual setting After selecting whether to use it in the popup window, click the Confirm button.

Setting up Security Group

You can configure the Security Group of an Auto-Scaling Group.

Reference
Even if you modify the Security Group, it does not apply to servers already created in the Auto-Scaling Group; it only applies to servers created thereafter. If you want to apply the updated Security Group to all servers in the Auto-Scaling Group, set the server count (Desired) to 0 to delete all existing servers, then adjust the server count (Desired) to the desired quantity again.

Follow these steps to configure the Security Group of an Auto-Scaling Group.

  1. All Services > Compute > Virtual Server Click the menu. 1. Navigate to the Service Home page of the Virtual Server.

  2. Click the Auto-Scaling Group menu. 2. Go to the Auto-Scaling Group List page.

  3. Auto-Scaling Group List page, click the resource to configure the Security Group. 3. Go to the Auto-Scaling Group Details page.

  4. Click the Edit button for Security Group. 4. Security Group Edit The popup window opens. 4. You can view the list of selectable Security Groups.

    CategoryDetailed description
    Security Group nameSecurity Group name
    Table. Security Group list items

  5. Security Group Edit In the popup window, after selecting the Security Group, click the Confirm button. 5. Security Group Modification Notification A popup window opens. 5. Security Group Modification Alert Check the message in the popup window and click the Confirm button.

Change status check

You can configure it to automatically replace servers that are in an Unhealthy state by checking the status of Virtual Server and Load Balancer.

Caution
  • The Load Balancer status check feature can be used only when a Load Balancer is in use.
  • A server that the user changes to Stop is not considered Unhealthy, so even when using the health check feature, the server will not be replaced.
To change whether the Load Balancer health check feature of an Auto-Scaling Group is enabled, follow these steps.

  1. All Services > Compute > Virtual Server Click the menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. Click the Auto-Scaling Group menu. 2. Go to the Auto-Scaling Group List page.
  3. On the Auto-Scaling Group List page, click the resource to modify the Load Balancer health check feature. 3. Navigate to the Auto-Scaling Group Details page.
  4. Click the Edit button of Status Check. 4. Edit Status Check The popup window opens.
  5. Status Check Edit In the popup, choose whether to enable Load Balancer Status Check.
  6. After entering the Grace period, click the Confirm button.
    • If you do not want to use the grace period, set the grace period to 0.

Auto-Scaling Group Manage Additional Information

You can set the Auto-Scaling Group’s Load Balancer to enabled and select the LB server group. For an Auto-Scaling Group that uses a Load Balancer, you can set it to unused.

Load Balancer Draining Timeout Edit

You can set the Load Balancer Draining Timeout of an Auto-Scaling Group.

Reference

Draining Timeout is the time to wait before removing a server from the Load Balancer.

  • Because sessions may remain connected to the server, setting a Draining Timeout and waiting can clean up sessions more safely.
  • If the Load Balancer is unused, the Draining Timeout cannot be set.
  • The default is 300 seconds, and it can be set from a minimum of 1 second up to a maximum of 3,600 seconds.

To set the Load Balancer Draining Timeout for an Auto-Scaling Group, follow these steps.

  1. All Services > Compute > Virtual Server menu, click it. 1. Navigate to the Service Home page of the Virtual Server.
  2. Click the Auto-Scaling Group menu. 2. Go to the Auto-Scaling Group List page.
  3. Auto-Scaling Group List page, click the resource to configure the Load Balancer Draining Timeout. 3. Auto-Scaling Group Details Navigate to the page.
  4. Load Balancer Click the tab. 4. Load Balancer Navigate to the list page.
  5. Click the Edit button of Draining Timeout. 5. Draining Timeout Edit A popup window opens.
  6. Edit Draining Timeout In the popup window, select whether to use Draining Timeout, and enter the Draining Timeout duration (seconds).
  7. Draining Timeout Edit After checking the input value in the popup window, click the Confirm button. 7. Draining Timeout Modification Notification A popup window opens. 7. Check the message in the notification popup and click the Confirm button.

Using Load Balancer

You can modify the Load Balancer of an Auto-Scaling Group. To configure the Load Balancer of an Auto-Scaling Group, follow these steps.

Reference
  • When a server in the Auto-Scaling Group is created, it is automatically added as a member of the selected Load Balancer’s LB server group, and when the server is terminated, it is removed from the LB server group’s members.
  • If Draining Timeout is in use, after waiting for the Draining Timeout (seconds), the server is removed from the LB server group’s members.
  • In the case of Member separation due to Load Balancer modification, it waits in Detach from LB state, and in the case of Member separation due to Scale In, it waits in Scale In state.
  1. All Services > Compute > Virtual Server Click the menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. Click the Auto-Scaling Group menu. 2. Navigate to the Auto-Scaling Group List page.
  3. Auto-Scaling Group List page, click the resource to use with Load Balancer. 3. Go to the Auto-Scaling Group Details page.
  4. Load Balancer Click the tab. 3. Load Balancer Navigate to the list page.
  5. Click the Edit button of the Load Balancer. 4. Load Balancer Edit The popup window opens.
  6. Edit Load Balancer Please select whether to use it in the popup window. 5. If you select Use, you can select a Load Balancer.
    CategoryDetailed description
    LB server groupLB server group name
    • Select the LB server group that is created in the selected VPC
    • LB server groups that use Weighted Round Robin or Weighted Least Connection load balancing cannot be selected
    PortPort information of the LB server group
    • When registering as a member of the LB server group, enter the required port information in the registration details.
    • Enter a value between 1 and 65,534
    Table. Load Balancer list items
    • Click the + button to add an LB server group. * A maximum of three is allowed. * You can remove the Load Balancer you added by clicking the X button.
  7. Check the Load Balancer list and click the Confirm button. 6. Load Balancer Modification Notification A popup window opens. 6. Check the message in the notification popup and click the Confirm button.
Caution
  • Please be aware that separating or connecting servers from the Load Balancer may impact the service.
  • If the Draining Timeout is in use, setting the Load Balancer to unused or removing some of the connected Load Balancers with the X button will not disconnect them immediately. * After waiting for the Draining Timeout (seconds), the server is detached from the Load Balancer. * At this point, the Auto-Scaling Group waits in the Detach from LB state.
Reference

Load Balancer Not Used

You can modify the Auto-Scaling Group’s Load Balancer to be disabled. To configure an Auto-Scaling Group without using a Load Balancer, follow these steps.

Caution
  • Please be aware that separating or connecting servers from the Load Balancer may impact the service.
  • If the Draining Timeout is in use, setting the Load Balancer to not used or removing some of the connected Load Balancers with the X button will not detach them immediately. * After waiting for Draining Timeout(seconds), the server is detached from the Load Balancer. * At this point, the Auto-Scaling Group waits in the Detach from LB state.
  1. All Services > Compute > Virtual Server Click the menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. Click the Auto-Scaling Group menu. 2. Auto-Scaling Group List Navigate to the page.
  3. On the Auto-Scaling Group List page, click the resource to set the Load Balancer to unused. 3. Auto-Scaling Group Details page.
  4. Load Balancer Click the tab. 4. Load Balancer Navigate to the list page.
  5. Click the Edit button of Load Balancer. 5. Load Balancer Edit The popup window opens.
  6. Edit Load Balancer Select whether to use it in the popup window. 6. If you deselect 사용, you will no longer use the Load Balancer.
  7. Uncheck the Use option and click the Confirm button. 7. Load Balancer Modification Notification A popup window opens. 7. Check the notification popup message and click the Confirm button.

Delete Auto-Scaling Group

Deleting unused Auto-Scaling Groups can reduce operating costs. However, if you terminate the Auto-Scaling Group, the running service may be stopped immediately, so you should proceed with the termination only after fully considering the impact of service interruption.

Caution
Please be aware that data cannot be recovered after deleting the service.

To terminate the Auto-Scaling Group, follow these steps.

  1. All Services > Compute > Virtual Server menu, click it. 1. Navigate to the Service Home page of the Virtual Server.
  2. Click the Auto-Scaling Group menu. 2. Go to the Auto-Scaling Group List page.
  3. Auto-Scaling Group List page, click the resource to terminate. 3. Go to the Auto-Scaling Group Details page.
  4. Click the Auto-Scaling Group Delete button.
  5. After the deletion is complete, verify on the Auto-Scaling Group List page that the resource has been deleted.

2.2.1 - Launch Configuration

To create an Auto-Scaling Group, a Launch Configuration must be created in advance.

Create Launch Configuration

You can create and use the Launch Configuration service in the Samsung Cloud Platform Console.

To create a Launch Configuration, follow these steps.

  1. All Services > Compute > Virtual Server menu, please click. 1. Navigate to the Service Home page of the Virtual Server.

  2. Click the Launch Configuration menu. 2. Go to the Launch Configuration List page.

  3. On the Launch Configuration List page, click the Create Launch Configuration button. 3. Go to the Launch Configuration creation page.

  4. On the Launch Configuration creation page, in the Image and version selection area, select the required information and click the Next button.

    Reference
    • Refer to the Virtual Server OS Image 제공 버전 for the images selectable in the Launch Configuration.
    • In Custom Image, you can select Image-Based and Snapshot-Based images.
      • Snapshot-Based type Image does not support future features such as distributed disk deployment, so for a stable Auto-Scaling configuration, we recommend using Image-Based type Image.
      • Image-Based type Image generation details can be found in Image-Based 유형의 Image 생성하기.

  5. Enter the required information in the Service Information Input area of the Launch Configurationp creation page.

    Category
    Required status
    Detailed description
    Launch Configuration nameEssentialLaunch Configuration name
    • a name to distinguish the Launch Configuration
    Service Type > Server TypeEssentialServer type of Launch Configuration
    • Standard: Standard specification commonly used
    • High Capacity: Large‑capacity server specification above Standard
    • Since supported Availability Zones may differ by server type, verification is required via Support Center > Contact Us
    Block StorageRequiredConfigure Block Storage according to the purpose of the Launch Configuration
    • Basic OS: The area where the OS is installed and used
      • Enter the size in Units; the minimum size varies depending on the OS Image type
        • Alma Linux: Enter a value between 2 and 1,536
        • Oracle Linux: Enter a value between 5 and 1,536
        • RHEL: Enter a value between 2 and 1,536
        • Rocky Linux: Enter a value between 2 and 1,536
        • Ubuntu: Enter a value between 1 and 1,536
        • Windows: Enter a value between 4 and 1,536
      • SSD: High‑performance general volume
      • HDD: General volume
      • SSD/HDD_KMS: Additional encrypted volume using Samsung Cloud Platform KMS (Key Management System) encryption keys
        • Encryption can be applied only at initial creation and cannot be changed afterward
        • Using the SSD_KMS disk type may result in performance degradation
    • Additional: Use when extra space beyond the OS area is needed
      • After selecting Use, enter the storage type and size
      • Click the + button to add storage, or the x button to delete (up to 25 can be added)
      • Enter the size in Units, with a value between 1 and 1,536
        • Since 1 Unit equals 8 GB, this creates 8 ~ 12,288 GB
      • SSD: High‑performance general volume
      • HDD: General volume
      • SSD/HDD_KMS: Additional encrypted volume using Samsung Cloud Platform KMS (Key Management System) encryption keys
        • Encryption can be applied only at initial creation and cannot be changed afterward
        • Using the SSD_KMS disk type may result in performance degradation
      • SSD_MultiAttach: Volume that can be attached to more than one server
    KeypairEssentialSelect the user authentication method for the Launch Configuration
    • Server authentication information for accessing the servers created by generating an Auto-Scaling Group with a Launch Configuration
    • New creation: Create a new one if a new Keypair is needed
    • Default login account list by OS
      • Alma Linux: almalinux
      • RHEL: cloud-user
      • Rocky Linux: rocky
      • Ubuntu: ubuntu
      • Windows: sysadmin
    File Storage SettingsSelectionWhen performing Scale-out/Scale-in, you can configure File Storage to be automatically attached
    • Volume: After clicking the Select button, choose the Volume to use in the File Storage selection popup
    • You can select up to a maximum of 5
    • Only Volumes intended as source that can be attached to a Virtual Server can be selected
      • High‑performance SSDs are disabled because they cannot be attached to a Virtual Server
    • The selected Volume can be removed by clicking the X button
    Table. Launch Configuration service information input items

  6. On the Launch Configuration Creation page, after entering information in the Additional Information Input area, click the Next button.

    Category
    required status
    Detailed description
    Init ScriptSelectionThe script that runs when a server in an Auto-Scaling Group using a Launch Configuration starts
    • Enter within 45,000 bytes
    • The Init Script must be a batch script for Windows, a shell script for Linux, or cloud‑init, depending on the selected image.
    tagSelectionAdd Tag
    • Up to 50 per resource can be added
    • Add Tag After clicking the Add Tag button, enter or select Key, Value values
    Table. Launch Configuration additional information entry fields

  7. On the Check Creation Information page, verify the entered information and the estimated amount, then click the Create button.

    • When creation is complete, verify the created Launch Configuration on the Launch Configuration list page.

Launch Configuration View detailed information

The Launch Configuration service lets you view and edit the complete list of resources and their detailed information. On the Launch Configuration Details page, it consists of Detailed Information, Tags, Operation History tabs.

To view the detailed information of the Launch Configuration, follow these steps.

  1. All Services > Compute > Virtual Server Click the menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. Click the Launch Configuration menu. 2. Go to the Launch Configuration list page.
  3. Launch Configuration List On the page, click the resource to view detailed information. 3. Launch Configuration Details Navigate to the page.
    • Launch Configuration Details At the top of the page, status information and additional feature information are displayed, and it consists of Details, Tags, Activity History tabs.
      CategoryDetailed description
      Launch Configuration statusThe status of the user-created Launch Configuration
      • Active: available
      Delete Launch ConfigurationDelete Launch Configuration button
      Table. Launch Configuration status information and additional features

Detailed Information

Launch Configuration List page allows you to view detailed information of the selected resource and, if necessary, modify the information.

CategoryDetailed description
serviceService name
Resource TypeResource Type
SRNUnique resource ID in Samsung Cloud Platform
  • In Launch Configuration, it means the Launch Configuration SRN
Resource nameresource name
  • In Launch Configuration, it refers to the Launch Configuration name
Resource IDUnique resource ID in the service
ConstructorUser who created the service
Creation Date/TimeService creation date and time
ModifierUser who edited the service information
Modification date and timeDate and time the service information was modified
Launch Configuration nameLaunch Configuration name
ImageWhen creating a Launch Configuration, the selected Image name
  • when creating servers in the Auto-Scaling Group that uses this Launch Configuration, the OS Image used
Auto-Scaling Group countNumber of Auto-Scaling Groups using a Launch Configuration
Server typeServer type set in the Launch Configuration
Block StorageBlock Storage information per server set in the Launch Configuration
  • type, capacity, and kind
KeypairServer authentication information set in the Launch Configuration
  • Keypair information to use when connecting to the servers of the Auto-Scaling Group that uses this Launch Configuration
Init ScriptInit Script set in the Launch Configuration
  • The script that runs when a server in an Auto Scaling Group using that Launch Configuration starts
File Storage SettingsFile Storage Volume name set in the Launch Configuration
  • Click the Volume name to view detailed File Storage information
Table. Launch Configuration detailed information tab items

Tag

On the Launch Configuration list page, you can view the tag information of the selected resource, and add, modify, or delete it.

CategoryDetailed description
Tag listTag list
  • You can view the Key and Value information of the tag
  • Up to 50 tags can be added per resource
  • When entering tags, you can search and select from the list of previously created Keys and Values
Table. Launch Configuration tag tab items

Operation History

Launch Configuration List page allows you to view the operation history of the selected resource.

CategoryDetailed description
Task History ListResource change history
  • Operation date/time, resource ID, resource name, operation details, event topic, operation result, operator information verification
Table. Launch Configuration Job History Tab Detailed Information Items

Launch Configuration Resource Management

Explains the resource management function of Launch Configuration.

Generate Image-Based type Image

To create an Image of the Image-Based type, follow the steps below.

information
  • If the image type is Snapshot-Based, it does not support future features such as distributed disk deployment. * We recommend using an Image with Image type set to Image-Based for a stable Auto-scaling configuration.
  • When generating an Image of the Image-Based type, only one OS area disk volume is included in the target. * Note that additionally attached data volumes are not included in the Image.
  • An Image can be created as an Image-Based type only if its Visibility is private and it has snapshot information.
  • If the OS disk size is 500 GB or more, you cannot create an Image of the Image-Based type.
  • Object Storage fees for Image storage will be billed.

To create an Image of the Image-Based type, follow the steps below.

  1. Click the All Services > Compute > Virtual Server menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. On the Service Home page, click the Image menu. 2. Go to the Image list page.
  3. On the Image List page, click the Image to create an Image of the Image-Based type. 3. Image Details go to the page.
  4. Create Image for Sharing button, click it. 4. A popup window opens to notify the creation of a shared Image. 4. The popup’s ‘Share to another Account’ content will not be processed. 4. If you only proceed with ‘Create Image for sharing’, an Image-Based type Image will be generated.
  5. After reviewing the contents of the popup window, click the Confirm button.

Delete Launch Configuration

Deleting unused Launch Configurations can reduce operating costs. However, terminating the Launch Configuration may cause the running service to stop immediately, so you should proceed with the termination only after fully considering the impact of service interruption.

Caution
Please be careful, as data cannot be recovered after deletion.

To delete a Launch Configuration, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. Click the Launch Configuration menu. 2. Go to the Launch Configuration List page.
  3. Launch Configuration List On the page, click the resource to be terminated. 3. Navigate to the Launch Configuration Details page.
  4. Click the Delete Launch Configuration button.
  5. After the deletion is complete, verify on the Launch Configuration list page that the resource has been deleted.
Caution
The Launch Configuration applied to the Auto-Scaling Group cannot be deleted. Please delete the Auto-Scaling Group first, then delete the Launch Configuration.

2.2.2 - Manage Policy

You can dynamically adjust the number of servers in an Auto-Scaling Group based on monitoring metrics. When the metric exceeds the threshold you set, the server count is adjusted. At that time, you can choose one of three methods to adjust the server count: increase or decrease by a specified number, increase or decrease by a specified percentage, or fix the server count to a given value. When servers are launched or terminated due to a policy, the monitoring metric, such as CPU utilization, may temporarily exceed the threshold registered in the policy. However, because this is a brief moment, a cooldown period is set to avoid treating it as an abnormal condition. You can add and manage policies for an Auto-Scaling Group created in the Samsung Cloud Platform Console.

Add Policy

You can add policies to an Auto-Scaling Group. To add a policy to an Auto-Scaling Group, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. Navigate to the Service Home page of Virtual Server.

  2. Click the Auto-Scaling Group menu. You will be taken to the Auto-Scaling Group List page.

  3. Auto-Scaling Group List page, click the resource you want to view detailed information for. You will be taken to the Auto-Scaling Group Details page.

  4. Click the Policy Tab. You will be taken to the Policy Tab page.

  5. Click the Add Policy button. The Add Policy popup window opens.

    Category
    Required
    Detailed description
    CategoryRequiredPolicy Category
    • Scale In: return server count
    • Scale Out: increase server count
    Policy NameRequiredName for policy-specific categorization
    Execution conditionsRequiredConditions to execute the policy
    • Statistic: How to calculate the Metric Type
      • Average: Average of servers in the Auto-Scaling Group
      • Min: Minimum value among servers in the Auto-Scaling Group
      • Max: Maximum value among servers in the Auto-Scaling Group
    • Metric Type: CPU Usage, Network In(bytes), Network Out(bytes), Network In(Packets), Network Out(Packets)
    • Operator: >= > <= <
    • Threshold: Threshold corresponding to the Metric Type
    • Period: Continuous duration required to trigger the condition (the condition must be continuously satisfied for N minutes for the policy to execute)
    execution unitRequiredPolicy execution method
    • Policy Type: Select the type of policy to execute.
      • Increase or decrease the number of servers by a specified count: Increase or decrease the server count to the Target Value
      • Increase or decrease the number of servers by a specified ratio: Increase or decrease by the Target Value ratio
      • Fix the number of servers to the entered value: Fix the server count according to the Target Value
    • Target Value: The number or ratio to apply for the selected Policy Type
    cooldownRequiredThe waiting time (seconds) when a server is started or terminated due to a policy
    • The default is 300 seconds, and it can be set between a minimum of 60 seconds and a maximum of 3,600 seconds.
    Table. Add Policy Popup Items
    Reference

    Policy > Cooldown Settings

    • When a server is started or terminated due to a policy, it waits for the configured cooldown period. Temporarily, the monitoring metric CPU utilization may exceed the threshold defined in the policy. However, because this is a transient moment rather than a condition for adjusting the number of servers, it is not considered an abnormal situation, and the system waits by applying the cooldown time.
    guide

    Policy execution operates within the configured Min/Max server count range.

    • Even if you input values outside the Min/Max server count range—such as increasing, returning, or fixing the number of servers—it operates within the configured Min/Max server count.
    • Example: When the minimum number of servers is 3, even if you set the server count fixed to 1, the server count does not drop to 1 and remains at the minimum of 3.

  6. Add Policy After entering the required values in the popup window, click the Confirm button. The added policy can be viewed in the Policy List.

Policy creation example

The following is an explanation of the policy example. Please refer to it when creating a policy.

Policy example description 1
CategoryExecution Conditionsexecution unitcooldown
Scale OutAverage CPU Usage >= 60% occurs for 1 minuteIncrease the server count by the specified number, incrementing by one unit.300 seconds
Table. Auto-Scaling Group policy example 1
  • If the average CPU usage of the servers in the Auto-Scaling Group exceeds 60% for one minute, a server is added one at a time.
  • When a server is added, the cooldown period is 300 seconds, and during this time, no server addition or removal due to policy occurs.
  • After the cooldown period ends, the policy execution conditions are checked again.
Policy example description 2
CategoryExecution Conditionsexecution unitcooldown
Scale InMin CPU Usage <= 5% occurs for 1 minuteScale the number of servers up or down by the specified ratio, returning 50%.300 seconds
Table. Auto-Scaling Group policy example 2
  • If the minimum CPU usage of servers in the Auto-Scaling Group remains below 5% for 1 minute, 50% of the current servers will be terminated.
  • When a server is terminated, the cooldown period is 300 seconds, and during this time, no server additions or removals occur due to policy.
  • After the cooldown period ends, the policy execution conditions are checked again.
Policy example description 3
CategoryExecution ConditionsExecution Unitcooldown
Scale OutMax CPU Usage >= 90% occurs for 1 minuteFix the number of servers to 5 based on the entered value300 seconds
Table. Auto-Scaling Group policy example 3
  • If the maximum CPU usage among the servers in the Auto-Scaling Group exceeds 90% for 1 minute and the current number of servers is less than 5, servers will be created up to a total of 5.
  • During server creation, the cooldown period is 300 seconds, and no server additions or removals due to policy occur during the cooldown period.
  • After the cooldown period ends, the policy execution conditions are checked again.

Edit Policy

You can modify the policies of an Auto-Scaling Group. To modify the policies of an Auto-Scaling Group, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. Navigate to the Service Home page of Virtual Server.

  2. Click the Auto-Scaling Group menu. Navigate to the Auto-Scaling Group List page.

  3. Auto-Scaling Group List page, click the resource to view detailed information. You will be taken to the Auto-Scaling Group Details page.

  4. Click the Policy Tab. You will be taken to the Policy Tab page.

  5. Click the More > Edit button for the policy you want to modify. The Policy Edit Popup opens.

    Category
    Required
    Detailed description
    CategoryRequiredPolicy Classification
    • Scale In: Return server count
    • Scale Out: Increase server count
    Policy NameRequiredName for policy-specific categorization
    Execution conditionsRequiredConditions to execute the policy
    • Statistic: Method to calculate the Metric Type
      • Average: Average of servers in the Auto-Scaling Group
      • Min: Minimum value among servers in the Auto-Scaling Group
      • Max: Maximum value among servers in the Auto-Scaling Group
    • Metric Type: CPU Usage, Network In(bytes), Network Out(bytes), Network In(Packets), Network Out(Packets)
    • Operator: >= > <= <
    • Threshold: Threshold corresponding to the Metric Type
    • Period: Continuous duration (in minutes) that triggers the execution condition
    execution unitRequiredPolicy execution method
    • Policy Type: Select the type of policy to execute.
      • Increase or decrease the number of servers by a specified count: Increase or reduce the server count to the Target Value
      • Increase or decrease the number of servers by a specified ratio: Increase or reduce by the Target Value ratio
      • Fix the number of servers to the entered value: Fix the server count according to the Target Value
    • Target Value: The number or ratio to execute the selected Policy Type
    Cool downRequiredThe waiting time (seconds) when a server is started or terminated due to a policy
    • The default is 300 seconds, and it can be set from a minimum of 1 second up to a maximum of 3,600 seconds
    Table. Policy edit popup items

  6. Edit Policy In the popup window, after entering the required values, click the Confirm button.

Policy addition and modification constraints

When adding or modifying a policy, constraints exist based on the policy classification, execution conditions, and the scope of those conditions. Below is an example of constraints for the policy. Refer to the constraint examples to add or modify the policy.

Example 1 - Need to check for duplicate registration of policy classification/execution conditions

Policy classification (Scale Out or Scale In) and execution condition (Metric type) cannot be registered redundantly with the same values.

Policy classificationPolicy NameExecution Condition(Statistic)Execution condition(Metric type)Execution condition range
Scale OutScaleOutPolicyAverageCPU Usage>= 60%
Table. Policy Constraint Example 1 - Pre-registration Policy

If a policy is registered as above, you cannot add a policy with the category (Scale Out) execution condition (Metric type=CPU Usage) or modify it to that condition.

Example 2 - Need to verify the execution condition range for execution conditions (Metric type) and execution conditions (Statistic) according to policy classification

When the policy type (Scale Out or Scale In) differs, duplicate registration of the execution condition range (Comparison operator + Threshold) is not allowed for the same execution condition (Metric type) and execution condition (Statistic).

Policy classificationPolicy NameExecution condition(Statistic)Execution condition(Metric type)Execution condition range
Scale OutScaleOutPolicyAverageCPU Usage>= 60%
Table. Policy Constraint Example 2 - Pre-registration Policy

If a policy is registered as shown above, you cannot add a policy as below or modify it under the following conditions.
If the average CPU Usage is 60% or higher, a Scale Out policy is already registered; therefore, registering a Scale In policy when the average CPU Usage is 60% or lower would duplicate the 60% execution condition and cannot be added.

Policy ClassificationPolicy NameExecution Condition(Statistic)Execution condition(Metric type)Execution condition range
Scale InAddUpdatePolicyAverageCPU Usage<= 60%
Table. Policy Constraint Example 2 - Non-Addable Policy

Example 3 - Need to verify the execution condition range for execution conditions (Metric type) and execution conditions (Statistic) according to policy classification

When the policy classification (Scale Out or Scale In) differs, duplicate registration of the execution condition range (Comparison operator + Threshold) is not allowed for the same execution condition (Metric type) and execution condition (Statistic).

Policy classificationPolicy NameExecution condition(Statistic)Execution condition (Metric type)Execution condition range
Scale InScaleInPolicyAverageCPU Usage<= 10%
Table. Policy Constraint Example 3 - Pre-registration Policy

If a policy is registered as shown above, you cannot add a policy as below or modify it under the following conditions.
When the average CPU Usage is 10% or less, a Scale In policy is already registered; therefore, if the average CPU Usage is less than 60% / 60% or less / 10% or more / exceeds 9%, you cannot register a Scale Out policy because the execution condition ranges would overlap.

Policy classificationPolicy NameExecution Condition(Statistic)Execution condition(Metric type)Execution condition range
Scale OutAddUpdatePolicy1AverageCPU Usage< 60%
Scale OutAddUpdatePolicy2AverageCPU Usage<= 60%
Scale OutAddUpdatePolicy3AverageCPU Usage>= 10%
Scale OutAddUpdatePolicy4AverageCPU Usage> 9%
Table. Policy constraint example 3 – non‑addable policy

Example 4 - Execution conditions (Metric type) and execution conditions (Statistic) can be registered according to the execution condition range based on policy classification

If the policy type (Scale Out or Scale In) differs, you can register when the execution condition (Metric type) is the same but the execution condition (Statistic) is different, or when the execution condition range (Comparison operator + Threshold) does not overlap.

Policy classificationPolicy NameExecution Condition(Statistic)Execution condition (Metric type)Execution Condition Range
Scale OutScaleOutPolicyAverageCPU Usage>= 60%
Table. Policy Constraint Example 4 - Pre-registration Policy

If a policy is already registered as shown above, you can add a policy as below or modify it with the following conditions. Registration is possible when the execution condition ranges do not overlap, or when the execution condition (Statistic) differs.

Policy classificationPolicy NameExecution Condition(Statistic)Execution condition (Metric type)Execution condition range
Scale InAddUpdatePolicy1AverageCPU Usage<= 10%
Scale InAddUpdatePolicy2MinCPU Usage<= 60%
Table. Policy constraint example 4 – Addable policy

Delete Policy

You can delete a policy of an Auto-Scaling Group. To delete a policy of an Auto-Scaling Group, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. Navigate to the Service Home page of Virtual Server.
  2. Click the Auto-Scaling Group menu. Navigate to the Auto-Scaling Group List page.
  3. Auto-Scaling Group List page: click the resource you want to view details for. You will be taken to the Auto-Scaling Group Details page.
  4. Click the Policy Tab. You will be taken to the Policy Tab page.
  5. Select the policy to delete and click the Delete button. The Policy Deletion Confirmation popup opens.
  6. Policy Deletion Confirmation Check the popup window and click the Confirm button.

2.2.3 - Manage Schedule

You can schedule reservations daily, weekly, monthly, or one-time, and set the desired number of servers at the specified time. This is useful when you can predict when you need to decrease or increase the number of servers.

Add schedule

You can add a schedule to an Auto-Scaling Group. To add a schedule to an Auto-Scaling Group, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. Navigate to the Service Home page of Virtual Server.
  2. Click the Auto-Scaling Group menu. Navigate to the Auto-Scaling Group List page.
  3. Auto-Scaling Group List page, click the resource you want to view detailed information for. You will be taken to the Auto-Scaling Group Details page.
  4. Click the Schedule Tab. You will be taken to the Schedule Tab page.
  5. Click the Add Schedule button. The Add Schedule popup window opens.
    Category
    required
    Detailed description
    Schedule nameRequiredName to distinguish by schedule
    Select server countRequiredSelect the number of servers to adjust when executing the schedule
    • Min: The minimum number of servers the Auto-Scailg Group should maintain
    • Desired: The target number of servers within the Auto-Scailg Group
    • Max: The maximum number of servers the Auto-Scailg Group can maintain
    Enter the number of serversRequiredEnter the value for the selected number of servers
    • Min value: Enter a value between 0 and 50. (Min≤Desired≤Max)
    • Desired value: Enter a value between 0 and 50. (Min≤Desired≤Max)
    • Max value: Enter a value between 0 and 50. (Min≤Desired≤Max)
    PeriodRequiredSchedule execution frequency
    • Daily: You can set the Start date, End date, and Permanent options for a daily schedule. You can also set the Time and Time zone
    • Weekly: You can set the Start date, End date, Permanent options, and the Time and Time zone. You can also select the Day of week on which the weekly schedule will run.
    • Monthly: You can set the Start date, End date, Permanent options, and the Time and Time zone. You can also enter the Date on which the monthly schedule will run
    • One-time: You can set the Time and Time zone. You can also set the Date on which the one-time schedule will run
    Start dateSelectionSet schedule start date
    • Cannot set a date earlier than the current date. The default is the current date.
    End dateSelectionSet schedule end date
    • Cannot set a date earlier than the current date. The default is the current date plus 7 days.
    permanentSelectionWhen configuring permanently, set the schedule end date to 9999-12-31
    timeRequiredSchedule execution time setting
    • Can be set in 30‑minute increments. Times earlier than the current date or current time cannot be set
    time zoneRequiredSchedule execution time zone (example: Asia/Seoul (GMT +09:00))
    day of the weekRequiredWhen Period is set to Weekly, select the Day of the week to execute the schedule
    DateRequired
    • When Period is set to Monthly, enter the Date on which the schedule will run
      • Enter one or more values from -31 to 31, excluding 0. (Example: 3,4,5)
    • When Period is set to Once, set the Date on which the schedule will run
      • Setting a date earlier than the current date is not allowed. The default is the current date.
    Table. Schedule addition popup items
  6. Add Schedule After entering the required values in the popup window, click the Confirm button.
  7. Add Schedule Confirmation After checking the message in the popup window, click the Confirm button.
Reference

If you select the schedule frequency as monthly, you must enter the schedule execution date, Date. Refer to the information below to register the schedule.

  • If you input a number greater than 0, it represents the day of the month.
    • Example: If you enter 1, you get August 1, September 1, …, December 1
  • If you enter a number less than 0, the calculation starts from the end of each month.
    • Entering -1 means the last day of each month.
      • Example: August 31, September 30, …, December 31
    • If you enter -2, it means the day before the last day of each month.
      • Example: August 30, September 29, …, December 30
    • Since the last day of each month varies—31, 30, 29, or 28 days—we allow calculation from the end of the month using negative numbers, as shown above, to handle schedules that need to run on the month’s final day.
Information
  • When the schedule runs, if the Min server count set in the schedule is greater than the Desired server count, or if the Max server count is less than the Desired server count, the Desired server count is also adjusted.
  • If there are schedules with overlapping execution times, they may not run correctly. Please try to avoid overlapping execution times as much as possible.

Modify schedule

You can modify the schedule of an Auto-Scaling Group. To modify the schedule of an Auto-Scaling Group, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. Navigate to the Service Home page of Virtual Server.
  2. Click the Auto-Scaling Group menu. Navigate to the Auto-Scaling Group List page.
  3. On the Auto-Scaling Group List page, click the resource for which you want to view detailed information. You will be taken to the Auto-Scaling Group Details page.
  4. Click the Schedule Tab. You will be taken to the Schedule Tab page.
  5. Click the More > Edit button of the schedule you want to modify. The Edit Schedule popup window will open.
    Category
    required
    Detailed description
    Schedule nameRequiredName to distinguish by schedule
    Select number of serversRequiredSelect the number of servers to adjust when executing the schedule
    • Min: The minimum number of servers the Auto-Scailg Group should maintain
    • Desired: The target number of servers within the Auto-Scailg Group
    • Max: The maximum number of servers the Auto-Scailg Group can maintain
    Enter number of serversRequiredEnter the value for the selected number of servers
    • Min value: Enter a value between 0 and 50. (Min≤Desired≤Max)
    • Desired value: Enter a value between 0 and 50. (Min≤Desired≤Max)
    • Max value: Enter a value between 0 and 50. (Min≤Desired≤Max)
    periodRequiredSchedule execution frequency
    • Daily: You can set the Start Date and End Date, and Permanent setting for the daily schedule. You can also set the Time and Time Zone
    • Weekly: You can set the Start Date and End Date, Permanent setting, and the Time and Time Zone. You can also select the Day of Week for the Weekly schedule to run.
    • Monthly: You can set the Start Date and End Date, Permanent setting, and the Time and Time Zone. You can also enter the Date for the Monthly schedule to run.
    • Once: You can set the Time and Time Zone. You can also set the Date for the Once schedule to run
    Start dateSelectionSet schedule start date
    • Cannot set a date earlier than the current date. The default is the current date.
    End dateSelectionSet schedule end date
    • Cannot set a date earlier than the current date. The default is the current date plus 7 days.
    PermanentSelectionWhen configuring permanently, set the schedule end date to 9999-12-31.
    timeRequiredSchedule execution time setting
    • Can be set in 30‑minute increments. Times earlier than the current date or current time cannot be set
    time zoneRequiredSchedule execution time zone (example: Asia/Seoul (GMT +09:00))
    day of the weekRequiredWhen Period is weekly, select the day of the week to run the schedule.
    DateRequired
    • When Period is set to Monthly, enter the Date on which the schedule will run
      • Enter one or more values from -31 to 31, excluding 0. (Example: 3,4,5)
    • When Period is set to Once, set the Date on which the schedule will run
      • Cannot set a date earlier than the current date. The default is the current date.
    Table. Schedule edit popup items
  6. Schedule Edit After entering the required values in the popup window, click the Confirm button.
  7. Schedule Modification Confirmation After checking the message in the popup window, click the Confirm button.

Delete schedule

You can delete the schedule of an Auto-Scaling Group. To delete the schedule of an Auto-Scaling Group, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. Navigate to the Service Home page of Virtual Server.
  2. Click the Auto-Scaling Group menu. You will be taken to the Auto-Scaling Group List page.
  3. On the Auto-Scaling Group List page, click the resource for which you want to view detailed information. You will be taken to the Auto-Scaling Group Details page.
  4. Click the Schedule Tab. You will be taken to the Schedule Tab page.
  5. Select the schedule to delete and click the Delete button. Schedule Deletion Confirmation popup will open.
  6. Schedule Deletion Confirmation Check the popup window and click the Confirm button.

2.2.4 - Manage Notification

You can designate a notification recipient and send alert messages via E‑mail or SMS for specific situations.

Reference
  • The notification method (E-mail or SMS) can be set on the Notification Settings page by selecting Notification Target as Service > Virtual Server Auto-Scaling.
  • For details on editing notification settings, refer to Edit Notification Settings.

Add notification

You can add notifications to an Auto-Scaling Group. To add notifications to an Auto-Scaling Group, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. Navigate to the Service Home page of Virtual Server.
  2. Click the Auto-Scaling Group menu. You will be taken to the Auto-Scaling Group List page.
  3. On the Auto-Scaling Group List page, click the resource to which you want to add notification information. You will be taken to the Auto-Scaling Group Details page.
  4. Click the Notification Tab. You will be taken to the Notification Tab page.
  5. Click the Add Notification button. The Add Notification popup opens.
  6. Add Notification After entering the required values in the popup window, click the Confirm button.
    CategoryDetailed description
    Notification timingNotification timing when an Auto-Scaling Group alert occurs
    • Server creation, Server termination, Server creation failure, Server termination failure, When policy execution conditions are met
    • Multiple selection allowed
    Notification recipientUser who will receive the notification when an alert occurs
    • Click the Add Notification Recipient button to select a user
    • Only Samsung Cloud Platform users can be selected as recipients
    Table. Notification items
Caution
When adding a notification recipient, verify that an email address exists and add it. Only users with login history (users who have registered an email or mobile phone number) can receive notifications.
  1. Add Notification Confirmation After checking the message in the popup window, click the Confirm button.

Edit Notification

You can modify the notification settings of an Auto-Scaling Group. To modify the notification settings of an Auto-Scaling Group, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. Navigate to the Service Home page of Virtual Server.
  2. Click the Auto-Scaling Group menu. Navigate to the Auto-Scaling Group List page.
  3. On the Auto-Scaling Group List page, click the resource to edit the notification information. You will be taken to the Auto-Scaling Group Details page.
  4. Click the Notification Tab. You will be taken to the Notification Tab page.
  5. In the notification list, click the More > Edit button for the notification you want to modify. The Edit Notification popup window will open.
  6. Edit Notification After editing the notification information in the popup window, click the Confirm button.
    CategoryDetailed description
    Notification timingNotification timing when an Auto-Scaling Group alert occurs
    • Server creation, Server termination, Server creation failure, Server termination failure, When policy execution conditions are met
    • Multiple selection allowed
    Table. Notification edit items
  7. Edit Notification Confirmation After checking the message in the popup window, click the Confirm button.

Delete Notification

You can delete notifications for an Auto-Scaling Group. To delete notifications for an Auto-Scaling Group, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. Navigate to the Service Home page of Virtual Server.
  2. Click the Auto-Scaling Group menu. You will be taken to the Auto-Scaling Group List page.
  3. On the Auto-Scaling Group List page, click the resource to edit the notification information. You will be taken to the Auto-Scaling Group Details page.
  4. Click the Notification Tab. You will be taken to the Notification Tab page.
  5. In the notification list, select the notification you want to delete, then click the Delete button. The Delete Notification Confirmation popup will open.
  6. Delete Notification Confirmation Review the popup window and click the Confirm button.

2.3 - API Reference

API Reference

2.4 - CLI Reference

CLI Reference

2.5 - Release Note

Virtual Server Auto-Scaling

2026.03.19
FEATURE Virtual Server and Load Balancer status check feature and other feature additions
  • Virtual Server and Load Balancer status check functionality has been added.
    • Check the status of Virtual Server and Load Balancer and automatically replace servers that are in an Unhealthy state.
  • You can attach File Storage to a Launch Configuration.
    • When creating a Launch Configuration, you can attach up to five File Storage Volumes.
  • You can monitor data by integrating with the ServiceWatch service.
  • When setting the execution conditions of the policy, Memory Usage was excluded from the Metric Type.
2025.10.23
FEATURE Virtual Server termination feature improvement
  • The termination functionality for Virtual Servers created by the Auto-Scaling Group has been improved.
    • When a Virtual Server is terminated, the Load Balancer is automatically detached.
    • The desired number of servers remains unchanged, and scale-out is performed according to the desired check batch.
2025.07.01
FEATURE Add new feature
  • Added a notification feature to Virtual Server Auto-Scaling.
    • You can add notification settings in the creation or detail screen of an Auto-Scaling Group.
  • You can set scaling policies when creating an Auto-Scaling Group.
  • Added the Metric Type to the Auto-Scaling Group policy.
    • Additional: Memory Usage, Network In(bytes), Network Out(bytes), Network In(Packets), Network Out(Packets)
  • You can set the Draining Timeout when connecting to a Load Balancer.
  • An Auto-Scaling Group can connect up to 50 Virtual Servers, and up to three LB server groups and ports.
2025.02.27
FEATURE Virtual Server Auto Scaling-Load Balancer Release of service integration and addition of NAT configuration feature
  • Virtual Server Auto-Scaling feature change
    • It will be released in conjunction with the Load Balancer service launching in February 2025.
    • NAT configuration functionality has been added to Auto-Scaling Group.
  • Samsung Cloud Platform Common Feature Change
    • Account, IAM, Service Home, tags, etc., have been updated to reflect common CX changes.
2024.11.19
NEW Virtual Server Auto Scaling Service Official Version Launch
  • Virtual Server Auto-Scaling creates an Auto-Scaling Group through a Launch Configuration and monitors and manages the servers.
  • Provides a scheduling method that allows setting the desired number of servers at a specified time, and a policy method that adjusts the number of servers based on CPU utilization.

3 - GPU Server

3.1 - Overview

Service Overview

GPU Server is a virtualized computing service that lets you freely allocate and use infrastructure resources such as CPU, GPU, and memory provided by the server, without having to purchase them individually, and allocate as much as needed at the required time. It is suitable for tasks that require fast computation speed, such as AI model experiments, predictions, and inference in a cloud environment, and you can flexibly select and use resources with optimized performance according to the type and scale of the work. The GPU Server provides the following features.

Provided features

  • GPU Server Management: Through a web-based console, users can directly Self Service create, delete, and modify GPU servers from provisioning to monitoring and billing.
  • Product offering by GPU quantity: Depending on the project’s purpose and scale, you can freely select the number of H100/A100 GPUs to configure a virtual server.
  • High‑Performance GPU Provision: We provide high‑performance GPU servers at physical‑server level using a pass‑through method.
  • Storage Connection: Provides additional attached storage besides the OS disk. * Block Storage, File Storage, Object Storage can be connected and used.
  • Strong Security Enforcement: Through the Security Group service, control inbound/outbound traffic exchanged with the external Internet or other VPC (Virtual Private Cloud) to securely protect the server.
  • Monitoring: You can view monitoring information such as the status of CPU, Memory, Disk, and GPU, which are computing resources, through the Cloud Monitoring service.
  • Network Configuration Management: The server’s subnet/IP can be conveniently changed from the values set at initial creation. * NAT IP provides a management feature that lets you enable or disable it as needed.
  • Key Pair method: To ensure a secure OS access method, we provide a Key Pair method instead of ID/PW login.
  • Image Management: You can create and manage Custom Images, and it provides sharing functionality between projects.
  • ServiceWatch Service Integration: You can monitor data through the ServiceWatch service.

Components

GPU Server provides GPUs, NVSwitch, and NVLink on top of virtualized computing resources.

caution
  • NVSwitch can be enabled and used only for instance types that allocate eight GPUs on a single GPU server.

Specifications by GPU Type

GPU (Graphic Processing Unit) performs the calculations required to generate the images that compose a computer screen, and because it is specialized for parallel processing, it can handle large amounts of data quickly, processing large‑scale parallel operations such as artificial intelligence (AI) and data analysis. The specifications of the GPU Types provided by the GPU Server service are as follows.

CategoryA100 TypeH100 TypeB300 Type
GPU ArchitectureNVIDIA AmpereNVIDIA HopperNVIDIA Blackwell Ultra
GPU Memory80 GiB80 GiB268 GiB
GPU Transistors54 billion 7N TSMC80 billion 4N TSMC208 billion 4NP TSMC
FP16 Tensor Core (Dense)312 TFLOPs989 TFLOPs2.25 PFLOPs
FP8 Tensor Core (Dense)Unsupported1,979 TFLOPs4.5 PFLOPs
FP4 Tensor Core (Dense)UnsupportedUnsupported13.5 PFLOPs
GPU Memory Bandwidth2,039 GB/s HBM2e3,352 GB/s HBM38 TB/s HBM3e
NVLink performanceNVLink 3NVLink 4NVLink 5
NVLink Signaling Rate25 GB/s (x12)25 GB/s (x18)50 GB/s (x18)
NVSwitch GPU-to-GPU bandwidth600 GB/s900 GB/s1.8 TB/s
Total NVSwitch aggregate bandwidth4.8 TB/s7.2 TB/s14.4 TB/s
Table. Specifications by GPU Type

NPU Type Specifications

The NPU (Network Processing Unit) is a processor specialized for AI inference operations, performing generative AI and various AI inference workloads based on high throughput and power efficiency. The specifications of the NPU Type provided by the GPU Server service are as follows.

CategoryFuriosa RNGD
ArchitectureTensor Contraction Processer
BF16256 TFLOPS
FP8512 TFLOPS
Memory BandwidthHBM3 1.5 TB/s
Memory CapacityHBM3 48 GB
Interconnect InterfacePCIe Gen5 x16
Table. Specifications by NPU Type

Server type

The server types offered by the GPU Server are as follows. For detailed information about the server types provided by GPU Server, see GPU Server 서버 타입.

CategoryServer typeCPU vCoreMemory(GB)GPU/NPU quantity
GPU-A100-1g1v16a1162341
GPU-A100-1g1v32a2324682
GPU-A100-1g1v64a4649364
GPU-A100-1g1v128a81281,8728
GPU-H100-2g2v12h1122341
GPU-H100-2g2v24h2244682
GPU-H100-2g2v48h4489364
GPU-H100-2g2v96h8961,8728
GPU-B300-3g3v16b1164801
GPU-B300-3g3v32b2329602
GPU-B300-3g3v64b4641,9204
GPU-B300-3g3v128b81283,8408
NPU-RNGD-1n1v8r181061
NPU-RNGD-1n1v16r2162122
NPU-RNGD-1n1v32r4324244
NPU-RNGD-1n1v64r8648488
Table. GPU Server server type

OS and driver version

The operating systems (OS) supported by the GPU Server are as follows. Please be aware that B300-type GPUs are supported only from a certain GPU version onward, so choose your image accordingly.

OSOS versionDriver versionServer type classification
Ubuntu24.04ND 580.126.20GPU-B300-3, GPU-H100-2, GPU-A100-1
Ubuntu24.04ND 570.195.03GPU-H100-2, GPU-A100-1
Ubuntu24.04FRD 2026.2.0NPU-RNGD-1
Ubuntu22.04ND 535.183.06GPU-H100-2, GPU-A100-1
RHEL9.6ND 580.126.20GPU-B300-3, GPU-H100-2, GPU-A100-1
RHEL8.1ND 580.126.20GPU-B300-3, GPU-H100-2, GPU-A100-1
RHEL8.1ND 535.183.06GPU-H100-2, GPU-A100-1
Table. GPU Server OS and driver version

Constraints

The GPU Server has the following constraints.

  • We are providing Ubuntu22.04 as the current OS version.
  • GPU is provided via Pass-through.
  • Physical GPU cards are created in units of 1, 2, 4, or 8 cards each. * –>

Preceding Service

This service must be installed in advance before creating it. Please prepare by referring to the user guide provided in advance.

Service CategoryserviceDetailed description
NetworkingVPCA service that provides an isolated virtual network in a cloud environment
NetworkingSecurity GroupVirtual firewall that controls server traffic
Table. GPU Server Preliminary Service

3.1.1 - Server type

GPU Server server type

GPU servers are categorized by the GPU type they offer, and the GPU used in a GPU server is determined by the server type selected when creating the GPU server. Please select the server type based on the specifications of the application you want to run on the GPU server.

The server types supported by the GPU Server are as follows.

GPU-H100-2 g2v12h1
Category
ExampleDetailed description
Server typeGPU-H100-2Provided server type classification
  • GPU-H100-2
    • GPU-H100 indicates the provided GPU type
    • 2 indicates the generation
  • GPU-A100-1
    • GPU-A100 indicates the provided GPU type
    • 1 indicates the generation
Server specificationsg2제공되는 서버 타입 구분 및 세대
  • g2
    • g는 GPU 서버 사양을 의미
    • 2은 세대를 의미
서버 사양v12vCore 개수
  • v2: 2개의 가상 코어
서버 사양h1GPU 종류와 수량
  • h1
    • h는 GPU-H100를 의미
    • 1은 GPU 1개를 의미
  • a2
    • a는 GPU-A100를 의미
    • 2은 GPU 2개를 의미
Table. GPU Server server type format

g1 server type

The g1 server type is a GPU Server that uses the NVIDIA A100 Tensor Core GPU, making it suitable for high‑performance applications.

  • Up to 8 NVIDIA A100 Tensor Core GPUs provided
  • Equipped with 6,912 CUDA cores and 432 Tensor cores per GPU.
  • Supports up to 128 vCPUs and 1,920 GB of memory
  • Maximum networking speed of 40 Gbps
  • 600 GB/s GPU and NVIDIA NVSwitch P2P communication
  • Additional AMD CPU-based server type provided (GPU-A100-1-A)
구분서버 타입GPUCPUMemoryGPU MemoryNetwork Bandwidth
GPU-A100-1g1v16a1116 vCore234 GB80 GiB최대 20 Gbps
GPU-A100-1g1v32a2232 vCore468 GB160 GiBMaximum 20 Gbps
GPU-A100-1g1v64a4464 vCore936 GB320 GiBUp to 40 Gbps
GPU-A100-1g1v128a88128 vCore1,872 GB640 GiBMaximum 40 Gbps
GPU-A100-1-Ag1av16a1116 vCore234 GB80 GiBMaximum 20 Gbps
GPU-A100-1-Ag1av32a2232 vCore468 GB160 GiBMaximum 20 Gbps
GPU-A100-1-Ag1av64a4464 vCore936 GB320 GiB최대 40 Gbps
GPU-A100-1-Ag1av128a88128 vCore1,872 GB640 GiB최대 40 Gbps
Table. GPU Server server type > GPU-A100-1 server type

g2 server type

The g2 server type is a GPU Server that uses the NVIDIA H100 Tensor Core GPU, making it suitable for high-performance applications.

  • Up to 8 NVIDIA H100 Tensor Core GPUs provided
  • Equipped with 16,896 CUDA cores and 528 Tensor cores per GPU
  • Supports up to 96 vCPUs and 1,920 GB of memory
  • Maximum networking speed of 40 Gbps
  • 900 GB/s GPU and NVIDIA NVSwitch P2P communication
구분서버 타입GPUCPUMemoryGPU MemoryNetwork Bandwidth
GPU-H100-2g2v12h1112 vCore234 GB80 GiBMaximum 20 Gbps
GPU-H100-2g2v24h2224 vCore468 GB160 GiBMaximum 20 Gbps
GPU-H100-2g2v48h4448 vCore936 GB320 GiBMaximum 40 Gbps
GPU-H100-2g2v96h8896 vCore1,872 GB640 GiBMaximum 40 Gbps
Table. GPU Server server type > GPU-H100-2 server type

g3 server type

The g3 server type is a GPU Server that uses the NVIDIA B300 Tensor Core GPU, making it suitable for high-performance applications.

  • Provides up to 8 NVIDIA B300 Tensor Core GPUs
  • Equipped with 20,480 CUDA cores and 640 Tensor cores per GPU.
  • Supports up to 128 vCPUs and 3,840 GB of memory.
  • Up to 40 Gbps networking speed
  • 1.8 TB/s GPU and NVIDIA NVSwitch P2P communication
CategoryServer typeGPUCPUMemoryGPU MemoryNetwork Bandwidth
GPU-B300-3g3v16b1116 vCore480 GB268 GiBMaximum 20 Gbps
GPU-B300-3g3v32b2232 vCore960 GB536 GiBMaximum 20 Gbps
GPU-B300-3g3v64b4464 vCore1,920 GB1,072 GiBUp to 40 Gbps
GPU-B300-3g3v128b88128 vCore3,840 GB2,144 GiB최대 40 Gbps
Table. GPU Server server type > GPU-B300-3 server type

n1 server type

The n1 server type is an NPU Server that uses the Furiosa RNGD NPU, making it suitable for AI inference workloads.

  • Provides up to 8 Furiosa RNGD NPUs
  • Supports 256 TFLOPS and 512 TFLOPS per NPU
  • Supports up to 64 vCPUs and 848 GB of memory
  • Maximum networking speed of 40 Gbps
  • PCIe Gen5 x16 interface support
구분서버타입NPUCPUMemoryNPU MemoryNetwork Bandwidth
NPU-RNGD-1n1v8r118 vCore106 GB48 GiBMaximum 20 Gbps
NPU-RNGD-1n1v16r2216 vCore212 GB96 GiBMaximum 20 Gbps
NPU-RNGD-1n1v32r4432 vCore424 GB192 GiBUp to 40 Gbps
NPU-RNGD-1n1v64r8864 vCore848 GB384 GiBMaximum 40 Gbps
Table. GPU Server server type > NPU-RNGD-1 server type

3.1.2 - Monitoring Metrics

Cloud Monitoring service termination notice

According to Samsung Cloud Platform’s policy, the Cloud Monitoring service is scheduled to be discontinued in September 2026.
Accordingly, after the September 2026 release, resource monitoring of the Samsung Cloud Platform via Cloud Monitoring will no longer be possible.

With the new alternative service, you can continuously perform resource monitoring by leveraging ServiceWatch released in October 2025.
ServiceWatch provides more modern and powerful features, replacing Cloud Monitoring to deliver a seamless monitoring environment.

If you are collecting metrics and logs through the Cloud Monitoring Agent, you need to switch to the ServiceWatch Agent.

For detailed information about ServiceWatch, please refer to ServiceWatch Overview.
Detailed information about ServiceWatch Agent can be found in the ServiceWatch Agent.

GPU Server Monitoring Metrics

The table below shows the monitoring metrics of the GPU server that can be viewed through Cloud Monitoring.

Even without installing the Agent, it provides basic monitoring metrics and the table below. Please check the GPU Server monitoring metrics (provided by default). Additionally, the metrics that can be viewed by installing the Agent are in the table. Please refer to the additional monitoring metrics for GPU Server (Agent installation required).

For detailed usage of Cloud Monitoring, refer to the Cloud Monitoring guide.

Performance Item NameExplanationunit
Memory Total [Basic]bytes of usable memorybytes
Memory Used [Basic]bytes of currently used memorybytes
Memory Swap In [Basic]bytes of the replaced memorybytes
Memory Swap Out [Basic]bytes of the replaced memorybytes
Memory Free [Basic]bytes of unused memorybytes
Disk Read Bytes [Basic]Read bytesbytes
Disk Read Requests [Basic]Number of read requestscnt
Disk Write Bytes [Basic]write bytesbytes
Disk Write Requests [Basic]Number of write requestscnt
CPU Usage [Basic]Average system CPU usage over 1 minute%
Instance State [Basic]Instance statusstate
Network In Bytes [Basic]Received bytesbytes
Network In Dropped [Basic]Incoming packet dropcnt
Network In Packets [Basic]Number of received packetscnt
Network Out Bytes [Basic]sent bytesbytes
Network Out Dropped [Basic]Transmit packet dropcnt
Network Out Packets [Basic]Number of transmitted packetscnt
Table. GPU Server Basic Monitoring Metrics (Provided by Default)
Performance Item NameExplanationunit
GPU CountNumber of GPUscnt
GPU Memory UsageMemory usage rate%
GPU Memory UsedMemory usageMB
GPU TemperatureGPU temperature
GPU Usageutilization%
GPU Usage [Avg]Overall average GPU utilization (%)%
GPU Power CapMaximum power capacity of the GPUW
GPU Power UsageCurrent GPU power usageW
GPU Memory Usage [Avg]GPU Memory Uti. AVG%
GPU Count in useNumber of GPUs in use by jobs on the nodecnt
Execution Status for nvidia-smiResult of running the nvidia-smi commandstatus
Core Usage [IO Wait]Ratio of CPU time spent in wait state (disk wait)%
Core Usage [System]Proportion of CPU time spent in kernel space%
Core Usage [User]Proportion of CPU time spent in user space%
CPU CoresNumber of CPU cores on the hostcnt
CPU Usage [Active]Percentage of CPU time used, excluding Idle and IOWait states%
CPU Usage [Idle]It is the proportion of CPU time spent in idle state.%
CPU Usage [IO Wait]The proportion of CPU time spent in a waiting state (disk wait).%
CPU Usage [System]Percentage of CPU time used by the kernel%
CPU Usage [User]Percentage of CPU time used in user space.%
CPU Usage/Core [Active]Percentage of CPU time used other than Idle and IOWait states%
CPU Usage/Core [Idle]It is the proportion of CPU time spent in idle state.%
CPU Usage/Core [IO Wait]This is the proportion of CPU time spent in a waiting state (disk wait).%
CPU Usage/Core [System]Percentage of CPU time used by the kernel%
CPU Usage/Core [User]Percentage of CPU time used in user space.%
Disk CPU Usage [IO Request]Proportion of CPU time during which I/O requests to the device were executed%
Disk Queue Size [Avg]The average queue length of requests executed for the device.num
Disk Read BytesThe number of bytes read per second from the device.bytes
Disk Read Bytes [Delta Avg]Average of system.diskio.read.bytes_delta for individual disksbytes
Disk Read Bytes [Delta Max]Maximum system.diskio.read.bytes_delta of individual disksbytes
Disk Read Bytes [Delta Min]Minimum system.diskio.read.bytes_delta of individual disksbytes
Disk Read Bytes [Delta Sum]Sum of the system.diskio.read.bytes_delta of individual disksbytes
Disk Read Bytes [Delta]Delta of the system.diskio.read.bytes value for each diskbytes
Disk Read Bytes [Success]Total number of bytes successfully read.bytes
Disk Read RequestsNumber of read requests to the disk device per secondcnt
Disk Read Requests [Delta Avg]Average of the system.diskio.read.count_delta for individual diskscnt
Disk Read Requests [Delta Max]Maximum system.diskio.read.count_delta for individual diskscnt
Disk Read Requests [Delta Min]Minimum of system.diskio.read.count_delta for individual diskscnt
Disk Read Requests [Delta Sum]Sum of system.diskio.read.count_delta for individual diskscnt
Disk Read Requests [Success Delta]Delta of system.diskio.read.count for each diskcnt
Disk Read Requests [Success]Total number of successful readscnt
Disk Request Size [Avg]The average size of requests executed on the device (unit: sectors).num
Disk Service Time [Avg]Average service time (milliseconds) of input requests executed on the device.ms
Disk Wait Time [Avg]Average time taken for requests executed on the supported device.ms
Disk Wait Time [Read]Average disk wait timems
Disk Wait Time [Write]Average disk wait timems
Disk Write Bytes [Delta Avg]Average of system.diskio.write.bytes_delta for each diskbytes
Disk Write Bytes [Delta Max]Maximum system.diskio.write.bytes_delta of individual disksbytes
Disk Write Bytes [Delta Min]Minimum of system.diskio.write.bytes_delta for individual disksbytes
Disk Write Bytes [Delta Sum]Sum of system.diskio.write.bytes_delta for individual disksbytes
Disk Write Bytes [Delta]Delta of the system.diskio.write.bytes value for each diskbytes
Disk Write Bytes [Success]Total number of bytes successfully written.bytes
Disk Write RequestsNumber of write requests to the disk device per secondcnt
Disk Write Requests [Delta Avg]Average of system.diskio.write.count_delta for individual diskscnt
Disk Write Requests [Delta Max]Maximum system.diskio.write.count_delta for individual diskscnt
Disk Write Requests [Delta Min]Minimum of system.diskio.write.count_delta for individual diskscnt
Disk Write Requests [Delta Sum]Sum of the system.diskio.write.count_delta of individual diskscnt
Disk Write Requests [Success Delta]Delta of system.diskio.write.count for each diskcnt
Disk Write Requests [Success]Total number of successful writescnt
Disk Writes BytesIt is the number of bytes per second written to the device.bytes
Filesystem Hang Checkfilesystem (local/NFS) hang check (normal:1, abnormal:0)status
Filesystem NodesIt is the total number of file nodes in the file system.cnt
Filesystem Nodes [Free]It is the total number of available file nodes in the file system.cnt
Filesystem Size [Available]Disk space (bytes) available to unauthorized usersbytes
Filesystem Size [Free]Available disk space (bytes)bytes
Filesystem Size [Total]Total disk space (bytes)bytes
Filesystem UsageUsed disk space percentage%
Filesystem Usage [Avg]Average of individual filesystem.used.pct values%
Filesystem Usage [Inode]inode usage%
Filesystem Usage [Max]Maximum among individual filesystem.used.pct%
Filesystem Usage [Min]minimum among individual filesystem.used.pct%
Filesystem Usage [Total]-%
Filesystem UsedUsed disk space (bytes)bytes
Filesystem Used [Inode]inode usagebytes
Memory FreeTotal amount of available memory (bytes).bytes
Memory Free [Actual]Actual usable memory (bytes).bytes
Memory Free [Swap]Available swap memory.bytes
Memory Totaltotal memorybytes
Memory Total [Swap]Total swap memory.bytes
Memory UsagePercentage of used memory%
Memory Usage [Actual]Percentage of memory actually used%
Memory Usage [Cache Swap]cached swap usage rate%
Memory Usage [Swap]Percentage of used swap memory%
Memory Usedused memorybytes
Memory Used [Actual]Actual memory used (bytes).bytes
Memory Used [Swap]Swap memory used.bytes
CollisionsNetwork collisioncnt
Network In BytesNumber of received bytesbytes
Network In Bytes [Delta Avg]Average of system.network.in.bytes_delta for individual networksbytes
Network In Bytes [Delta Max]Maximum of system.network.in.bytes_delta for each networkbytes
Network In Bytes [Delta Min]Minimum system.network.in.bytes_delta for each networkbytes
Network In Bytes [Delta Sum]Sum of system.network.in.bytes_delta for individual networksbytes
Network In Bytes [Delta]Delta of received byte countbytes
Network In DroppedNumber of deleted packets among incoming packetscnt
Network In ErrorsNumber of errors during receptioncnt
Network In PacketsNumber of received packetscnt
Network In Packets [Delta Avg]Average of system.network.in.packets_delta for each networkcnt
Network In Packets [Delta Max]Maximum of system.network.in.packets_delta for each networkcnt
Network In Packets [Delta Min]Minimum of system.network.in.packets_delta for individual networkscnt
Network In Packets [Delta Sum]Sum of system.network.in.packets_delta for individual networkscnt
Network In Packets [Delta]Delta of received packet countcnt
Network Out BytesNumber of transmitted bytesbytes
Network Out Bytes [Delta Avg]Average of system.network.out.bytes_delta for each networkbytes
Network Out Bytes [Delta Max]Maximum of system.network.out.bytes_delta for individual networksbytes
Network Out Bytes [Delta Min]Minimum of system.network.out.bytes_delta for individual networksbytes
Network Out Bytes [Delta Sum]Sum of system.network.out.bytes_delta for individual networksbytes
Network Out Bytes [Delta]Delta of transmitted byte countbytes
Network Out DroppedNumber of deleted packets among outgoing packets.cnt
Network Out ErrorsNumber of errors during transmissioncnt
Network Out PacketsNumber of transmitted packetscnt
Network Out Packets [Delta Avg]Average of system.network.out.packets_delta for each networkcnt
Network Out Packets [Delta Max]Maximum of system.network.out.packets_delta for each networkcnt
Network Out Packets [Delta Min]Minimum of system.network.out.packets_delta for each networkcnt
Network Out Packets [Delta Sum]Sum of system.network.out.packets_delta for individual networkscnt
Network Out Packets [Delta]Delta of transmitted packet countcnt
Open Connections [TCP]All open TCP connectionscnt
Open Connections [UDP]All open UDP connectionscnt
Port UsageAvailable port usage rate%
SYN Sent SocketsNumber of sockets in SYN_SENT state (when connecting from local to remote)cnt
Kernel PID Maxkernel.pid_max valuecnt
Kernel Thread Maxkernel.threads-max valuecnt
Process CPU UsagePercentage of CPU time consumed by the process since the last update.%
Process CPU Usage/CorePercentage of CPU time used by the process since the last event.%
Process Memory UsageProportion of main memory (RAM) occupied by a process%
Process Memory UsedResident Set size. The amount of memory a process occupies in RAM.bytes
Process PIDprocess pidPID
Process PPIDparent process PIDPID
Processes [Dead]Number of dead processescnt
Processes [Idle]Number of idle processescnt
Processes [Running]Number of running processescnt
Processes [Sleeping]Number of sleeping processescnt
Processes [Stopped]stopped processes countcnt
Processes [Total]Total number of processescnt
Processes [Unknown]Number of processes with an unsearchable or unknown statuscnt
Processes [Zombie]Zombie processes countcnt
Running Process Usageprocess usage rate%
Running ProcessesNumber of running processescnt
Running Thread UsageThread usage rate%
Running ThreadsTotal number of threads running in running processescnt
Context Switchescontext switch count (per second)cnt
Load/Core [1 min]The load over the last 1 minute divided by the number of corescnt
Load/Core [15 min]The load over the last 15 minutes divided by the number of corescnt
Load/Core [5 min]The load over the last 5 minutes divided by the number of corescnt
Multipaths [Active]External storage connection path status = active countcnt
Multipaths [Failed]External storage connection path status = failed countcnt
Multipaths [Faulty]External storage connection path status = faulty countcnt
NTP Offsetmeasured offset of the last sample (the time difference between the NTP server and the local environment)num
Run Queue LengthExecution queue lengthnum
UptimeOS uptime (milliseconds).ms
Context SwitchiesCPU context switch count (per second)cnt
Disk Read Bytes [Sec]Number of bytes read from a Windows logical disk in 1 secondcnt
Disk Read Time [Avg]Average data read time (seconds)sec
Disk Transfer Time [Avg]Disk average wait timesec
Disk UsageDisk usage%
Disk Write Bytes [Sec]Number of bytes written in one second on a Windows logical diskcnt
Disk Write Time [Avg]Average data write time (seconds)sec
Pagingfile UsagePaging file usage%
Pool Used [Non Paged]Nonpaged Pool usage in kernel memorybytes
Pool Used [Paged]Paged Pool usage in kernel memorybytes
Process [Running]Number of currently running processescnt
Threads [Running]Number of currently running threadscnt
Threads [Waiting]Number of threads waiting for processor timecnt
Table. Additional monitoring metrics for GPU Server (Agent installation required)

3.1.3 - ServiceWatch Metrics

The GPU Server sends metrics to ServiceWatch. The metrics provided by default monitoring are data collected at 5‑minute intervals. If detailed monitoring is enabled, you can view data collected at 1‑minute intervals.

information
  • The basic and detailed monitoring of the GPU Server are provided with the same metrics as the Virtual Server, and the namespace is also provided as Virtual Server.
  • GPU-related metrics are provided through ServiceWatch Agent, and for instructions on collecting metrics using ServiceWatch Agent, refer to the ServiceWatch Agent guide.
Reference
For how to view metrics in ServiceWatch, refer to the ServiceWatch guide.

Refer to How-to guides > ServiceWatch Detailed Monitoring Activation for how to enable detailed monitoring of the GPU Server.

Basic Metrics

The following are the basic metrics for the Virtual Server namespace.

The indicators whose names are displayed in bold below are the key indicators selected among the basic indicators provided by Virtual Server. The key metrics are used to build service dashboards that are automatically created for each service in ServiceWatch.

Each metric indicates through the user guide which statistical value is meaningful to view for that metric, and among the meaningful statistics, the statistical values shown in bold text are the primary statistics. In the service dashboard, you can view primary metrics using the primary statistical values.

Performance itemsDetailed descriptionunitmeaningful statistics
Instance StateInstance status display
  • 1 - Active
  • 0 - Off
None
  • Total
CPU UsageCPU usagePercent
  • Average
  • Maximum
  • Minimum
Disk Read BytesBytes read from block device (bytes)Bytes
  • Total
  • Average
  • Maximum
  • Minimum
Disk Read RequestsNumber of read requests on a block deviceCount
  • Total
  • Average
  • Maximum
  • Minimum
Disk Write BytesWrite capacity (bytes) on block deviceBytes
  • Total
  • Average
  • Maximum
  • Minimum
Disk Write RequestsNumber of write requests on block deviceCount
  • Total
  • Average
  • Maximum
  • Minimum
Network In BytesReceived bytes (capacity) on the network interfaceBytes
  • Total
  • Average
  • Maximum
  • Minimum
Network In DroppedNumber of packet drops received on the network interfaceCount
  • Total
  • Average
  • Maximum
  • Minimum
Network In PacketsNumber of packets received on the network interfaceCount
  • Total
  • Average
  • Maximum
  • Minimum
Network Out BytesData transmitted on the network interface (bytes)Bytes
  • Total
  • Average
  • Maximum
  • Minimum
Network Out DroppedNumber of packet drops transmitted from the network interfaceCount
  • Total
  • Average
  • Maximum
  • Minimum
Network Out PacketsNumber of packets transmitted on the network interfaceCount
  • Total
  • Average
  • Maximum
  • Minimum
Table. Virtual Server Basic Metrics

3.2 - How-to guides

Users can create the service by entering the required GPU Server information and selecting detailed options through the Samsung Cloud Platform Console.

Create GPU Server

You can create and use a GPU Server service from the Samsung Cloud Platform Console.

To create a GPU Server, follow these steps.

  1. Click the All Services > Compute > GPU Server menu. 1. Navigate to the Service Home page of the GPU Server.

  2. On the Service Home page, click the Create GPU Server button. 2. GPU Server Creation Go to the page.

  3. GPU Server Creation page, input the information required to create the service, and select detailed options.

    • Select the required information in the Image and Version Selection area.
      Category
      required status
      Detailed description
      ImageRequiredSelect the provided Image type
      • Standard: Samsung Cloud Platform standard Image
        • RHEL, Ubuntu
      • Custom: User-created Image
      • Kubernetes: Kubernetes Image
        • Ubuntu
      Image versionRequiredSelect the version of the chosen Image
      • Provide a list of versions for the offered server Image
      Table. GPU Server image and version selection input fields
    • Enter or select the required information in the Service Information Input area.
      Category
      required status
      Detailed description
      Number of serversRequiredNumber of GPU Server instances to create concurrently
      • only numeric input allowed, enter a value between 1 ~ 100
      Service Type > Server TypeRequiredGPU Server server type
      • Indicates the specifications of a GPU-type server, and select a server that includes 1, 2, 4, or 8 GPUs
      Service Type > Planned ComputeSelectionStatus of resources with Planned Compute configured
      • In Use: Number of resources with Planned Compute that are currently in use
      • Configured: Number of resources with Planned Compute configured
      • Coverage preview: Amount applied per resource by Planned Compute
      • Apply for Planned Compute service: Navigate to the Planned Compute service application page
      Block StorageRequiredSet the Block Storage used by the GPU Server according to its purpose
      • Default: Area where the OS is installed and used
        • Capacity can be entered in Unit increments (minimum capacity varies by OS image type)
          • RHEL: values between 3 and 1,536 are allowed
          • Ubuntu: values between 3 and 1,536 are allowed
        • SSD: high‑performance general volume
        • HDD: general volume
        • SSD/HDD_KMS: additional encrypted volume that uses Samsung Cloud Platform KMS (Key Management System) encryption keys
          • Encryption can be applied only at initial creation (cannot be changed after creation)
          • Performance degradation may occur when using the SSD_KMS disk type
        • SSD_Provisioned: SSD volume with configurable IOPS and Throughput
      • Additional: used when extra space beyond the OS area is needed
        • After selecting Use, enter the storage type and capacity
        • To add storage, click the + button (up to 25 can be added); to delete, click the x button
        • Capacity can be entered as a value between 1 and 1,536 in Unit increments
          • Since 1 Unit equals 8 GB, 8 ~ 12,288 GB can be created
        • SSD: high‑performance general volume
        • HDD: general volume
        • SSD/HDD_KMS: additional encrypted volume that uses Samsung Cloud Platform KMS (Key Management System) encryption keys
          • Encryption can be applied only at initial creation (cannot be changed after creation)
          • Performance degradation may occur when using the SSD_KMS disk type
        • HDD/SSD_MultiAttach: volume that can be attached to two or more servers
        • SSD_Provisioned: SSD volume with configurable IOPS and Throughput
      • Delete on termination: If Delete on Termination is set to Use, the volume will be terminated together with the server
        • Volumes that have snapshots are not deleted even when Delete on termination is set to Use
        • A multi‑attach volume can be deleted only when the server being removed is the last remaining server attached to the volume
      Max IOPSRequiredEnter the maximum IOPS value between 5,000 and 20,000
      • Disk type can be set only if it is SSD_Provisioned
      Max ThroughputRequiredEnter the maximum Throughput value between 250~1,000
      • Can be set only when the disk type is SSD_Provisioned
      Table. GPU Server Service Configuration Items
    • In the Required Information Input area, enter or select the required information.
      Category
      required status
      Detailed description
      Server nameRequiredIf the number of selected servers is 1, enter a name to distinguish the server
      • Set the hostname to the entered server name
      • Enter within 63 characters using English letters, numbers, spaces, and special characters (- _)
      Server name PrefixRequiredEnter a Prefix to distinguish each server generated when the selected number of servers is 2 or more
      • Automatically generated as the user input value (prefix) + ‘-#’ format
      • Enter within 59 characters using English letters, numbers, spaces, and special characters (-, _)
      Network Settings > Create New Network PortRequiredSet the network where the GPU Server will be installed
      • Select a pre-created VPC.
      • General Subnet: Select a pre-created general Subnet
        • IP can be set to auto-generate or user input, and if input is selected, the user can directly enter the IP
        • NAT: Available only when there is a single server and the VPC is connected to an Internet Gateway. Checking Use allows selection of a NAT IP
        • NAT IP: Select a NAT IP
          • If there is no NAT IP to select, click the Create New button to generate a Public IP
          • Refresh button to view and select the created Public IP
          • Creating a Public IP incurs charges according to the Public IP pricing policy
      • Local Subnet (optional): Select Use for the local Subnet
        • It is not a required element for creating the service
        • A pre-created local Subnet must be selected
        • IP can be set to auto-generate or user input, and selecting Input allows the user to directly enter the IP
      Network Settings > Assign Existing Network PortRequiredSet the network where the GPU Server will be installed
      • Select a pre‑created VPC
      • General Subnet: Select a pre‑created general Subnet and Port
        • NAT: Available only when there is a single server and the VPC is connected to an Internet Gateway; checking the option allows you to select a NAT IP.
        • NAT IP: Select a NAT IP
          • If no NAT IP is available to select, click the Create New button to generate a Public IP
          • Refresh button to view and select the created Public IP
      • Local Subnet (Optional): Select Use for the local Subnet
        • Select a pre‑created local Subnet and Port
      Security GroupSelectionSettings required to connect to the server
      • Selection: You can select up to 5 pre‑created Security Groups
      • Create New: If there is no Security Group to apply, create one separately in the Security Group service
      • If you do not configure a Security Group, all connections will be blocked, so you must set it to allow required access
      KeypairRequiredUser authentication method to use when connecting to the server
      • Default login account list by OS
        • RHEL: cloud-user
        • Ubuntu: ubuntu
      Table. GPU Server required information input items
    • Additional Information Input area, enter or select the required information.
      Category
      required status
      Detailed description
      LockSelectionLock usage setting
      • When using Lock, it prevents actions such as server termination, start, and stop from being executed, thereby avoiding malfunctions caused by mistakes
      Init scriptSelectionScript executed when the server starts
      • The init script must be written as a Batch script for Windows, a Shell script for Linux, or cloud‑init, depending on the image type.
      • Up to 45,000 bytes can be entered
      tagSelectionAdd Tag
      • Up to 50 per resource can be added
      • After clicking the Add Tag button, enter or select Key, Value values
      Table. GPU Server Additional Information Input Items
  4. Summary Check the detailed information and estimated charges generated in the panel, and click the Create button.

  5. When the popup notifying creation opens, click the Confirm button.

    • Once creation is complete, check the created resources on the GPU Server List page.

Check GPU Server detailed information

The GPU Server service allows you to view and edit the full resource list and detailed information. The GPU Server Details page consists of Details, Tags, Activity Log tabs. To view detailed information about the GPU Server service, follow these steps.

  1. All Services > Compute > GPU Server Click the menu. 1. Navigate to the Service Home page of the GPU Server.
  2. Click the GPU Server menu on the Service Home page. 2. GPU Server List Go to the page.
  3. GPU Server List page, click the resource to view detailed information. 3. Navigate to the GPU Server Details page.
    • GPU Server Details page displays status information and additional feature information, and consists of Details, Tags, Activity History tabs.
    • For detailed information about GPU Server Additional Features, please refer to GPU Server 관리 부가 기능.
      CategoryDetailed description
      GPU Server statusStatus of the GPU Server created by the user
      • Build: State where the Build command has been received
      • Building: Build in progress
      • Networking: Server creation process in progress
      • Scheduling: Server creation process in progress
      • Block_Device_Mapping: Connecting Block Storage during server creation
      • Spawning: State where the server creation process is ongoing
      • Active: Available state
      • Powering_off: State when a stop request is made
      • Deleting: Server deletion in progress
      • Reboot_Started: Reboot in progress state
      • Error: Error state
      • Migrating: State where the server is migrating to another host
      • Reboot: State where the Reboot command has been received
      • Rebooting: Reboot in progress
      • Rebuild: State where the Rebuild command has been received
      • Rebuilding: State when a Rebuild request is made
      • Rebuild_Spawning: State where the Rebuild process is ongoing
      • Shutoff: State when Powering off is completed
      Server controlButtons to change server status
      • Start: Start a stopped server
      • Stop: Stop a running server
      • Restart: Restart a running server
      Image generationCreate a user custom image from the current server’s image
      Console logView the console log of the current server
      Create dumpCreate a dump of the current server
      • The dump file is created inside the GPU Server
      RebuildAll data and settings of the existing server are deleted, and a new server is set up
      Service cancellationCancel service button
      Table. GPU Server status information and additional features
Notice
When using the mig feature, you must recheck the mig settings after the GPU Server’s Rebooting state has finished.

Detailed Information

GPU Server List page allows you to view detailed information of the selected resource and, if needed, modify the information.

CategoryDetailed description
serviceservice name
Resource TypeResource Type
SRNUnique resource ID in Samsung Cloud Platform
  • In the GPU Server service, it refers to the GPU Server SRN
Resource nameResource name
  • In the GPU Server service, it refers to the GPU Server name
Resource IDUnique resource ID in the service
ConstructorUser who created the service
Creation date and timeService creation date and time
ModifierUser who edited the service information
Modification date and timeDate and time the service information was modified
Server nameserver name
Server typeDisplay vCPU, memory, GPU information
  • If you need to change to a different server type, click the Edit button to configure
Image nameOS image and version of the service
LockDisplay lock usage status
  • If you need to change the Lock property value, click the Edit button to set it
Keypair nameUser-configured server authentication information
Planned ComputeResource status with Planned Compute configured
LLM EndpointURL for using LLM
ServiceWatch detailed monitoringWhen enabled, data monitoring is available in the ServiceWatch service
  • You can set the activation status by clicking the Edit button
NetworkNetwork information of the GPU Server
  • VPC name, standard Subnet name, IP, NAT IP, NAT IP status, Security Group name
  • If you need to change the NAT IP value, click the Edit button to configure
  • If you need to change the Security Group, click the Edit button to configure
  • Add as new network: select a standard Subnet and IP
    • You can select another standard Subnet within the same VPC
    • IP can be set to auto‑generate or manual entry; if manual is selected, the user can directly input the IP
  • Add using existing port: select a pre‑created standard Subnet and port
Local SubnetLocal Subnet information of GPU Server
  • Local Subnet name, Local Subnet IP, Security Group name
  • If a Security Group change is required, you can click the Edit button to configure
  • Add as new network: select Local Subnet and IP
    • You can select a different local Subnet within the same VPC
    • IP can be set to auto-generated or user input, and if Input is selected, the user enters the IP directly
  • Add with existing port: select a pre-created local Subnet and port
Block StorageInformation of Block Storage attached to the server
  • Volume ID, Volume Name, Type, Capacity, Connection Info, Category, Delete on termination, Status
  • Add: Additional Block Storage can be attached if needed
  • Edit Delete on termination: Modify the Delete on termination value
  • More > Detach: Detach the selected Block Storage from the list
Table. GPU Server Detailed Information Tab Items

Caution
When using ServiceWatch Detailed Monitoring, additional charges will apply.

Tag

GPU Server list page allows you to view the tag information of the selected resource, and to add, modify, or delete it.

CategoryDetailed description
Tag listTag list
  • You can view the tag’s Key, Value information
  • Up to 50 tags can be added per resource
  • When entering tags, you can search and select from the list of previously created Keys and Values
Table. GPU Server tag tab items

Job History

GPU Server List page allows you to view the operation history of the selected resource.

CategoryDetailed description
Task History ListResource change history
  • Operation timestamp, resource ID, resource name, operation details, event topic, operation result, check operator information
Table. Work History Tab Detailed Information Items

Control GPU Server Operation

If you need to control the operation of the created GPU Server resources, you can perform the task from the GPU Server List or GPU Server Details page. You can start, stop, or restart a running server.

Getting Started with GPU Server

You can start a GPU server that is shut off. To start the GPU Server, follow these steps.

  1. All Services > Compute > GPU Server Click the menu. 1. Navigate to the Service Home page of the GPU Server.
  2. Click the GPU Server menu on the Service Home page. 2. GPU Server List Go to the page.
  3. GPU Server List page, click the resource to start among the shutoff servers, and navigate to the GPU Server Details page.
    • On the GPU Server list page, you can start each resource via the right More button.
    • After selecting multiple servers with the checkboxes, you can control multiple servers simultaneously using the Start button at the top.
  4. On the GPU Server Details page, click the Start button at the top to start the server. 4. In the Status Indicator item, check the status of the changed server.
    • When the GPU Server startup is complete, the server status changes from Shutoff to Active.
    • For detailed information about the GPU Server status, please refer to GPU Server 상세 정보 확인하기.

Stop GPU Server

You can stop a GPU Server that is running (Active). To stop the GPU Server, follow the steps below.

  1. All Services > Compute > GPU Server Click the menu. 1. Navigate to the Service Home page of the GPU Server.
  2. Click the GPU Server menu on the Service Home page. 2. GPU Server List Go to the page.
  3. On the GPU Server List page, click the resource to stop among the servers that are active (Active), and go to the GPU Server Details page.
    • On the GPU Server List page, you can stop each resource using the right More button.
    • After selecting multiple servers with the checkboxes, you can control multiple servers simultaneously using the Stop button at the top.
  4. On the GPU Server Details page, click the Stop button at the top to start the server. 4. Check the status of the changed server in the Status Indicator item.
    • When the GPU Server shutdown is complete, the server status changes from Active to Shutoff.
    • For detailed information about the GPU Server status, refer to GPU Server 상세 정보 확인하기.

Restart GPU Server

You can restart the created GPU Server. To restart the GPU server, follow these steps.

  1. All Services > Compute > GPU Server Click the menu. 1. Navigate to the Service Home page of the GPU Server.
  2. Click the GPU Server menu on the Service Home page. 2. GPU Server List Go to the page.
  3. GPU Server List page, click the resource to restart, and navigate to the GPU Server Details page.
    • On the GPU Server List page, you can restart each resource through the right More button.
    • After selecting multiple servers with the checkboxes, you can control multiple servers simultaneously using the Restart button at the top.
  4. On the GPU Server Details page, click the Restart button at the top to start the server. 4. Check the status of the changed server in the Status Indicator item.
    • During a GPU Server restart, the server status goes through Rebooting and finally changes to Active.
    • For detailed information about the GPU Server status, please refer to GPU Server 상세 정보 확인하기.

GPU Server Resource Management

If you need server control and management functions for the created GPU Server resources, you can perform tasks on the GPU Server Resource List or GPU Server Details page.

Create Image

You can create an image of a running GPU server.

Reference

This document provides instructions on how to create a user custom image from the image of a running GPU server.

  • GPU Server List or GPU Server Details page, click the Create Image button to generate a user Custom Image.

To create an Image of the GPU Server, follow the steps below.

  1. All Services > Compute > GPU Server Click the menu. 1. Navigate to the Service Home page of the GPU Server.

  2. Click the GPU Server menu on the Service Home page. 2. GPU Server List Go to the page.

  3. On the GPU Server List page, click the resource to create an Image. 3. Navigate to the GPU Server Details page.

  4. On the GPU Server Details page, click the Create Image button. 4. Navigate to the Image creation page.

    • Enter the required information in the Service Information Input area.
      Category
      required status
      Detailed description
      Image nameRequiredName of the image to be created
      • English letters, numbers, spaces, and special characters(- _) using them, enter within 200 characters
      Table. Image service information input items
  5. Check the input information and click the Generate button.

    • Once creation is complete, check the created resources on the All Services > Compute > GPU Server > Image List page.
Notice
  • When an Image is created, the generated Image is stored in the Object Storage used as internal storage. * Therefore, a Object Storage usage fee is charged for Image storage.
  • Since the file system of an image generated from an active GPU server cannot be guaranteed to be intact, it is recommended to stop the server before creating the image.

Enable detailed monitoring for ServiceWatch

By default, the GPU Server is linked to the basic monitoring of the ServiceWatch and Virtual Server namespaces. You can enable detailed monitoring as needed to more quickly identify operational issues and take action. For detailed information about ServiceWatch, refer to the ServiceWatch 개요.

Reference
GPU Server provides basic and detailed monitoring in the same namespace as Virtual Server. GPU Server’s GPU metrics will be provided by the ServiceWatch Agent. (Scheduled for December 2025)
Caution
Basic monitoring is provided for free, but enabling detailed monitoring incurs additional charges. Please note the usage.

To enable detailed ServiceWatch monitoring on the GPU Server, follow these steps.

  1. All Services > Compute > GPU Server menu, click it. 1. Navigate to the Service Home page of the GPU Server.
  2. Click the GPU Server menu on the Service Home page. 2. GPU Server List Go to the page.
  3. On the GPU Server list page, click the resource to enable detailed ServiceWatch monitoring. 3. Navigate to the GPU Server Details page.
  4. On the GPU Server Details page, click the Edit button for ServiceWatch detailed monitoring. 4. ServiceWatch Detailed Monitoring Edit Navigate to the popup window.
  5. ServiceWatch Detailed Monitoring Edit In the popup window, after selecting Enable, review the instructions and click the Confirm button.
  6. Check the ServiceWatch detailed monitoring items on the GPU Server Details page.

Disable detailed monitoring of ServiceWatch

Caution
Disabling detailed monitoring is required for cost efficiency. Maintain detailed monitoring only when absolutely necessary, and disable detailed monitoring otherwise.

To disable detailed monitoring of ServiceWatch on the GPU Server, follow these steps.

  1. All Services > Compute > GPU Server Click the menu. 1. Go to the Service Home page of the GPU Server.
  2. Click the GPU Server menu on the Service Home page. 2. GPU Server List Go to the page.
  3. On the GPU Server List page, click the resource to disable ServiceWatch detailed monitoring. 3. Navigate to the GPU Server Details page.
  4. GPU Server Details page, click the Edit button for ServiceWatch detailed monitoring. 4. ServiceWatch Detailed Monitoring Edit Navigate to the popup window.
  5. ServiceWatch Detailed Monitoring Edit In the popup window, after deselecting Enable, review the guidance message and click the Confirm button.
  6. Check the detailed ServiceWatch monitoring items on the GPU Server Details page.

GPU Server management additional features

For managing the GPU Server, you can view Console logs, generate Dumps, and perform Rebuilds. To view the Console logs, create a Dump, and perform a Rebuild on the GPU Server, follow these steps.

Check console log

You can view the current console log of the GPU Server.

To check the console log of the GPU Server, follow these steps.

  1. All Services > Compute > GPU Server Click the menu. 1. Navigate to the Service Home page of the GPU Server.
  2. Click the GPU Server menu on the Service Home page. 2. GPU Server List Go to the page.
  3. On the GPU Server List page, click the resource to view the console log. 3. Navigate to the GPU Server Details page.
  4. On the GPU Server Details page, click the Console Log button. 4. Console Log navigates to the popup window.
  5. Console Log Check the console log displayed in the popup window.

Create Dump

To create a dump file on the GPU Server, follow these steps.

  1. All Services > Compute > GPU Server Click the menu. 1. Navigate to the Service Home page of the GPU Server.
  2. Click the GPU Server menu on the Service Home page. 2. GPU Server List Go to the page.
  3. GPU Server List page, click the resource to view detailed information. 3. Navigate to the GPU Server Details page.
  4. On the GPU Server Details page, click the Generate Dump button.
    • The dump file is created inside the GPU server.

Execute Rebuild

You can delete all data and settings of the existing GPU Server and rebuild it on a new server.

To perform a Rebuild of the GPU Server, follow these steps.

  1. All Services > Compute > GPU Server Click the menu. 1. Navigate to the Service Home page of the GPU Server.
  2. Click the GPU Server menu on the Service Home page. 2. GPU Server List Go to the page.
  3. On the GPU Server List page, click the resource to perform Rebuild. 3. Navigate to the GPU Server Details page.
  4. On the GPU Server Details page, click the Rebuild button.
    • During a GPU Server Rebuild, the server status changes to Rebuilding, and when the Rebuild is complete, it returns to the state it was in before the Rebuild.
    • For detailed information about the GPU Server status, please refer to GPU Server 상세 정보 확인하기.

Terminate GPU Server

If you terminate an unused GPU server, you can reduce operating costs. However, if you terminate the GPU server, the running service may be stopped immediately, so you should proceed with the termination only after fully considering the impact of service interruption.

Caution
Please be aware that data cannot be recovered after terminating the service.

To cancel the GPU Server, follow the steps below.

  1. All Services > Compute > GPU Server Click the menu. 1. Navigate to the Service Home page of the GPU Server.
  2. Click the GPU Server menu on the Service Home page. 2. GPU Server List Go to the page.
  3. On the GPU Server List page, select the resource to cancel, and click the Cancel Service button.
    • The termination of attached storage depends on the Delete on termination setting, so refer to 해지 제약 사항.
  4. When termination is complete, check on the GPU Server List page whether the resources have been terminated.

Termination constraints

When a GPU Server termination request cannot be processed, a popup will provide guidance. Please refer to the case below.

Cancellation not possible
  • When an LB server group is attached: Terminate the attached LB server group first.
  • If Lock is set: Please change the Lock setting to disabled and try again.

The termination of attached storage depends on the Delete on termination setting.

Delete on termination per setting
  • Whether the volume is deleted also varies depending on the Delete on termination setting.
    • Delete on termination When not set: Even if you terminate the GPU Server, the volume will not be deleted.
    • Delete on termination When enabled: terminating the GPU Server will delete the associated volume.
  • Volumes that have a snapshot will not be deleted even if Delete on termination is set.
  • A multi-attach volume can be deleted only when the server being deleted is the last remaining server attached to the volume.

3.2.1 - Manage Image

Users can create the service by entering the required information for the Image service within the GPU Server service and selecting detailed options through the Samsung Cloud Platform Console.

Create Image

You can create an image of a running GPU Server. To create an image of a GPU Server, please refer to Image Creation.

Check Image detailed information

Image service allows you to view and edit the full resource list and detailed information. The Image Details page consists of Detailed Information, Tags, Operation History tabs.

To view detailed information of the Image service, follow these steps.

  1. Click the All Services > Compute > GPU Server menu. You will be taken to the Service Home page of GPU Server.
  2. On the Service Home page, click the Image menu. You will be taken to the Image list page.
  3. On the Image List page, click the resource to view detailed information. You will be taken to the Image Detail page.
    • Image Details page displays status information and additional feature information, and consists of Details, Tags, Activity Log tabs.
      CategoryDetailed description
      Image statusStatus of user-created Image
      • Active: Available state
      • Queued: Image has been uploaded and is waiting for processing after creation
      • Importing: Image has been uploaded and is currently being processed after creation
      Share with another accountImage can be shared with another Account
      • The Image’s Visibility must be set to Shared in order to be shared with another Account
      Delete imageButton to delete the Image
      • Once the Image is deleted, it cannot be restored
      Table. GPU Server Image status information and additional features

Detailed Information

Image list page lets you view detailed information of the selected resource and modify it if necessary.

CategoryDetailed description
serviceService name
Resource TypeResource Type
SRNUnique resource ID in Samsung Cloud Platform
  • refers to the SRN of a GPU Server Image
Resource nameImage name
Resource IDImage ID
constructorUser who created the Image
Creation date and timeImage creation timestamp
editorUser who edited the Image
Modification dateImage modification timestamp
image nameImage name
Minimum diskMinimum disk capacity (GB) of the Image
  • If you need to modify the minimum disk, click the Edit button to set it
Minimum RAMMinimum RAM size (GB) of the Image
OS typeOS type of the image
OS hash algorithmOS hash algorithm method
VisibilityDisplay access permissions for the image
  • Private can be used only within the project, and Shared can be shared across projects
ProtectedSelect whether image deletion is prohibited
  • Checking Use can prevent accidental deletion of images
  • This setting can be changed after the image is created
image file URLImage file URL uploaded when generating an image
  • GPU Server detail page does not display images created through the image generation menu
Sharing statusCurrent status of sharing images with another Account
  • Approved Account ID: ID of the Account for which sharing is approved
  • Modification date and time: The date and time when sharing was requested to another Account; if the sharing status changes from Pending to Accepted, it is updated to that date and time
  • Status: Approved status
    • Accepted: sharing is approved and active
    • Pending: awaiting approval
  • Stop sharing: sharing has been stopped
  • Select the Account ID to stop sharing from the list, then click the Stop sharing button at the top of the list to stop sharing all at once
Table. Image detailed information tab items

tag

On the Image List page, you can view the tag information of the selected resource and add, modify, or delete it.

CategoryDetailed description
Tag listTag list
  • You can view the Key and Value information of the tag
  • Up to 50 tags can be added per resource
  • When entering a tag, you can search and select from the list of previously created Keys and Values
Table. Image tag tab items

Job History

You can view the operation history of the selected resource on the Image List page.

CategoryDetailed description
Task History ListResource Change History
  • Operation Time, Resource ID, Resource Name, Operation Details, Event Topic, Operation Result, Check Operator Information
Table. GPU Server Image Job History Tab Detailed Information Items

Image Resource Management

Describes the control and management functions of the generated Image.

Share to another Account

To share an Image with another Account, follow the steps below.

  1. Log in to the account to be shared and click the All Services > Compute > GPU Server menu. Go to the GPU Server’s Service Home page.
  2. On the Service Home page, click the Image menu. You will be taken to the Image List page.
  3. On the Image List page, click the Image you want to control. You will be taken to the Image Details page.
  4. Click the Share to another Account button. Navigate to the Share image to another Account page.
    • Share with another Account feature allows you to share an Image with another Account. To share an Image with another Account, the Image’s Visibility must be Shared.
  5. Share image to another Account page, enter the required information, and click the Done button.
    Category
    Required
    Detailed description
    image name-Name of the image to share
    • Input not allowed
    Image ID-Shareable image ID
    • Input not allowed
    Shared Account IDRequiredEnter another Account ID to share
    • English letters, numbers, special characters- within 64 characters
    Table. Required input fields for sharing images to another Account
  6. You can view the information in the sharing status of the Image Details page.
    • When the request is first made, the status is Pending, and it changes to Accepted once approval is completed by the account receiving the share.
Notice
Only images created by uploading an image file from the current user can be shared with another Account. If you create a Custom Image from the image of a running GPU Server, it cannot be shared with another Account, and this feature will be provided in the future, so please note.

Receive sharing from another Account

To receive an Image shared from another Account, follow these steps.

  1. Log in to the account to be shared and click the All Services > Compute > GPU Server menu. Navigate to the GPU Server’s Service Home page.
  2. On the Service Home page, click the Image menu. You will be taken to the Image List page.
  3. On the Image List page, click the Receive Image Share button. You will be taken to the Receive Image Share popup.
  4. Receive Image Sharing In the popup window, enter the Image’s resource ID you want to receive, and click the Confirm button.
  5. When image sharing is complete, you can view the shared Image in the Image list.

Delete Image

You can delete unused Images. However, since an Image cannot be recovered after deletion, you should carefully consider the impact before performing the deletion.

Caution
Please note that data cannot be recovered after deleting the service.

To delete the Image, follow these steps.

  1. Click the All Services > Compute > GPU Server menu. Go to the GPU Server’s Service Home page.
  2. On the Service Home page, click the Image menu. You will be taken to the Image List page.
  3. Image list page, select the resource to delete, and click the Delete button.
    • On the Image List page, select multiple Image check boxes and click the Delete button at the top of the resource list.
  4. After deletion is complete, verify on the Image list page that the resource has been removed.

3.2.2 - Manage Keypair

Users can create the service by entering the required Keypair information within the GPU Server service and selecting detailed options through the Samsung Cloud Platform Console.

Create a Keypair

You can create and use the Keypair service while using the GPU Server service in the Samsung Cloud Platform Console.

To create a keypair, follow these steps.

  1. Click the All Services > Compute > GPU Server menu. You will be taken to the Service Home page of GPU Server.
  2. On the Service Home page, click the Keypair menu. You will be taken to the Keypair List page.
  3. On the Keypair List page, click the Keypair Create button. You will be taken to the Keypair Create page.
    • Enter the required information in the Service Information Input area.
      Category
      Required
      Detailed description
      Keypair nameRequiredEnter the name of the Keypair to create
      • using English letters, numbers, spaces, and special characters (-, _) within 255 characters
      Keypair typeRequiredssh
      Table. Keypair service information input fields
    • Additional Information Input area, please enter or select the required information.
      Category
      Required status
      Detailed description
      tagSelectionAdd Tag
      • Up to 50 can be added per resource
      • After clicking the Add Tag button, enter or select Key, Value values
      Table. Keypair additional information input fields
      Caution
      • After creation is complete, you can download the Key only once. Since reissuance is not possible, make sure it has been downloaded.
      • Store the downloaded Private Key in a safe place.
  4. Check the input information and click the Create button.
    • After creation is complete, check the created resources on the Keypair List page.

View detailed information of the Keypair

The Keypair service allows you to view and edit the full resource list and detailed information. Keypair Details page consists of Details, Tags, Activity Log tabs.

To view detailed information about a keypair, follow these steps.

  1. Click the All Services > Compute > GPU Server menu. You will be taken to the Service Home page of GPU Server.
  2. On the Service Home page, click the Keypair menu. You will be taken to the Keypair List page.
  3. On the Keypair List page, click the resource to view its details. You will be taken to the Keypair Details page.
    • Keypair Details page displays status information and additional feature information, and consists of Details, Tags, Activity Log tabs.

Detailed Information

Keypair List page allows you to view detailed information of the selected resource and edit the information if needed.

CategoryDetailed description
serviceService name
Resource TypeResource Type
SRNUnique resource ID in Samsung Cloud Platform
  • In Keypair, it refers to the Keypair SRN
Resource nameKeypair name
Resource IDKeypair’s unique resource ID
constructorUser who created the keypair
Creation date and timeKeypair creation timestamp
editorUser who modified the keypair information
Modification dateDate and time the keypair information was modified
Keypair nameKeypair name
FingerprintA unique value for identifying the key
User IDUser ID of the keypair creator
public keyPublic key information
Table. Keypair detailed information tab items

tag

On the Keypair List page, you can view the tag information of the selected resource, and add, modify, or delete it.

CategoryDetailed description
Tag listTag list
  • You can view the Key and Value information of the tag
  • Up to 50 tags can be added per resource
  • When entering a tag, you can search and select from the list of previously created Keys and Values
Table. Keypair Tag Tab Items

Job History

On the Keypair List page, you can view the operation history of the selected resource.

CategoryDetailed description
Task History ListResource Change History
  • Operation Time, Resource ID, Resource Name, Operation Details, Event Topic, Operation Result, Check operator information
Table. Keypair operation history tab detailed information items

Keypair Resource Management

Describes the control and management functions of a keypair.

Get public key

To retrieve the public key, follow these steps.

  1. Click the All Services > Compute > GPU Server menu. You will be taken to the Service Home page of GPU Server.

  2. On the Service Home page, click the Keypair menu. You will be taken to the Keypair List page.

  3. On the Keypair List page, click the More button at the top and then click the Import Public Key button. You will be taken to the Import Public Key page.

    • Enter or select the required information in the Required Information Input area.
      Category
      Required
      Detailed description
      Keypair nameRequiredName of the Keypair to create
      Keypair typeRequiredssh
      public keyRequiredEnter public key
      • Load file: Select the Attach file button to attach the public key file
        • Only files with the following extension (.pem) can be attached
      • Enter public key: Paste the copied public key value
        • The public key value can be copied from the Keypair Details page
      Table. Required input fields for retrieving the public key
  4. Review the entered information and click the Complete button.

    • Once creation is complete, check the created resources on the Keypair List page.

Delete Keypair

You can delete unused Keypairs. However, once a Keypair is deleted it cannot be recovered, so please review the impact thoroughly beforehand before proceeding with deletion.

Caution
Please note that data cannot be recovered after deleting the service.

To delete a keypair, follow these steps.

  1. Click the All Services > Compute > GPU Server menu. Go to the Service Home page of GPU Server.
  2. On the Service Home page, click the Keypair menu. You will be taken to the Keypair List page.
  3. On the Keypair List page, select the resource to delete, and click the Delete button.
  • On the Keypair List page, select multiple Keypair check boxes and click the Delete button at the top of the resource list.
  1. After deletion is complete, check the Keypair List page to confirm that the resource has been removed.

3.2.3 - Use Multi-instance GPU on GPU Server

After creating a GPU Server, you can enable the MIG (Multi-instance GPU) feature on the GPU Server’s VM (Guest OS) and create an instance for use.

NVIDIA Multi-instance GPU Introduction

NVIDIA Multi-instance GPU (hereafter referred to as MIG) supports safely partitioning a GPU into GPU instances and running CUDA applications starting with the NVIDIA Ampere architecture. Through this, multiple users can each utilize different GPU resources to achieve optimal GPU utilization. This feature is especially useful for workloads that do not fully utilize the GPU’s computing capacity, and users can run multiple workloads in parallel to maximize utilization.

Using Multi-instance GPU feature

To use the MIG feature, create an NVIDIA GPU Server on the Samsung Cloud Platform, then enable and disable MIG. The order of applying and removing MIG is as follows.

MIG application order
Enable MIG → Create GPU Instance → Create Compute Instance → Use MIG
MIG release order
Delete Compute Instance → Delete GPU Instance → Disable MIG feature (deactivate)
Reference
  • MIG can be used on Samsung Cloud Platform’s next-generation GPU Server or MNGC (Multi-node GPU Cluster).
  • For system requirements to use MIG, refer to the NVIDIA Multi-Instance GPU User Guide.

Applying and Using MIG

After activating the MIG and creating an Instance to assign tasks, the tasks proceed in the following order.

MIG application order
Enable MIG → Create GPU Instance → Create Compute Instance → Use MIG
Note
The example of applying MIG is explained based on an A100 GPU server.

Activate MIG

  1. Check the GPU status on the VM Instance (GuestOS) before applying MIG.

    • Check whether MIG mode is Disabled.
      Color mode
      $ nvidia-smi
      Mon Sep 27 08:37:08 2021
      +-----------------------------------------------------------------------------+
      | NVIDIA-SMI 470.57.02    Driver Version: 470.57.02    CUDA Version: 11.4 |
      | -------------------------------+----------------------+---------------------- |
      | GPU  Name        Persistence-M | Bus-Id        Disp.A | Volatile Uncorr. ECC |
      | Fan  Temp  Perf  Pwr:Usage/Cap | Memory-Usage | GPU-Util  Compute M. |
      |  |  | MIG M. |
      | ===============================+======================+====================== |
      | 0  NVDIA A100-SXM...  Off | 00000000:05:00.0 Off | 0 |
      | N/A   32C   P0    59W / 400W | 0MiB / 81251MiB | 0%      Default |
      |  |  | Disabled |
      +-------------------------------+----------------------+----------------------+
      
      +-----------------------------------------------------------------------------+
      | Processes: |
      | GPU   GI   CI       PID   Type   Process name                   GPU Memory |
      | ID   ID                                                   Usage |
      | ============================================================================= |
      | No running processes found |
      +-----------------------------------------------------------------------------+
      $ nvidia-smi
      Mon Sep 27 08:37:08 2021
      +-----------------------------------------------------------------------------+
      | NVIDIA-SMI 470.57.02    Driver Version: 470.57.02    CUDA Version: 11.4 |
      | -------------------------------+----------------------+---------------------- |
      | GPU  Name        Persistence-M | Bus-Id        Disp.A | Volatile Uncorr. ECC |
      | Fan  Temp  Perf  Pwr:Usage/Cap | Memory-Usage | GPU-Util  Compute M. |
      |  |  | MIG M. |
      | ===============================+======================+====================== |
      | 0  NVDIA A100-SXM...  Off | 00000000:05:00.0 Off | 0 |
      | N/A   32C   P0    59W / 400W | 0MiB / 81251MiB | 0%      Default |
      |  |  | Disabled |
      +-------------------------------+----------------------+----------------------+
      
      +-----------------------------------------------------------------------------+
      | Processes: |
      | GPU   GI   CI       PID   Type   Process name                   GPU Memory |
      | ID   ID                                                   Usage |
      | ============================================================================= |
      | No running processes found |
      +-----------------------------------------------------------------------------+
      Code block. nvidia-smi command - Check GPU disabled status (1)
      Color mode
      $ nvidia-smi –L
      GPU 0: NVIDIA A100-SXM-80GB (UUID: GPU-c956838f-494a-92b2-6818-56eb28fe25e0)
      $ nvidia-smi –L
      GPU 0: NVIDIA A100-SXM-80GB (UUID: GPU-c956838f-494a-92b2-6818-56eb28fe25e0)
      Code block. nvidia-smi command - Check GPU disabled status (2)
  2. Enable MIG (Enable) for each GPU on the VM Instance (GuestOS) and reboot the VM Instance.

    Color mode
    $ nvidia-smi –I 0 –mig 1
    Enabled MIG mode for GPU 00000000:05:00.0
    All done.
    
    # reboot
    $ nvidia-smi –I 0 –mig 1
    Enabled MIG mode for GPU 00000000:05:00.0
    All done.
    
    # reboot
    Code block. nvidia-smi command - enable MIG

Reference

When using a GPU and configuring MIG, you may encounter the following warning message. If the warning appears, check whether any programs are running on the GPU.

Warning: MIG mode is in pending enable state for GPU 00000000:05:00.0: In use by another client. 00000000:05:00.0 is currently being used by one or more other processes (e.g. CUDA application or a monitoring application such as another instance of nvidia-smi).
  1. Check the GPU status after applying MIG on the VM Instance(GuestOS).
    • Check whether MIG mode is Enabled.
      Color mode
      $ nvidia-smi
      Mon Sep 27 09:44:33 2021
      +-----------------------------------------------------------------------------+
      | NVIDIA-SMI 470.57.02    Driver Version: 470.57.02    CUDA Version: 11.4 |
      | -------------------------------+----------------------+---------------------- |
      | GPU  Name        Persistence-M | Bus-Id        Disp.A | Volatile Uncorr. ECC |
      | Fan  Temp  Perf  Pwr:Usage/Cap | Memory-Usage | GPU-Util  Compute M. |
      |  |  | MIG M. |
      | ===============================+======================+====================== |
      | 0  NVDIA A100-SXM...  Off | 00000000:05:00.0 Off | On |
      | N/A   32C   P0    59W / 400W | 0MiB / 81251MiB | 0%      Default |
      |  |  | Enabled |
      +-------------------------------+----------------------+----------------------+
      +-----------------------------------------------------------------------------+
      | MIG devices: |
      +-----------------------------------------------------------------------------+
      | GPU  GI  CI  MIG | Memory-Usage | Vol | Shared |
      | ID  ID  Dev | BAR1-Usage | SM     Unc | CE  ENC  DEC  OFA  JPG |
      |  |  | ECC |  |
      | ============================================================================= |
      | No MIG devices found |
      +-----------------------------------------------------------------------------+
      +-----------------------------------------------------------------------------+
      | Processes: |
      | GPU   GI   CI       PID   Type   Process name                   GPU Memory |
      | ID   ID                                                   Usage |
      | ============================================================================= |
      | No running processes found |
      +-----------------------------------------------------------------------------+
      $ nvidia-smi
      Mon Sep 27 09:44:33 2021
      +-----------------------------------------------------------------------------+
      | NVIDIA-SMI 470.57.02    Driver Version: 470.57.02    CUDA Version: 11.4 |
      | -------------------------------+----------------------+---------------------- |
      | GPU  Name        Persistence-M | Bus-Id        Disp.A | Volatile Uncorr. ECC |
      | Fan  Temp  Perf  Pwr:Usage/Cap | Memory-Usage | GPU-Util  Compute M. |
      |  |  | MIG M. |
      | ===============================+======================+====================== |
      | 0  NVDIA A100-SXM...  Off | 00000000:05:00.0 Off | On |
      | N/A   32C   P0    59W / 400W | 0MiB / 81251MiB | 0%      Default |
      |  |  | Enabled |
      +-------------------------------+----------------------+----------------------+
      +-----------------------------------------------------------------------------+
      | MIG devices: |
      +-----------------------------------------------------------------------------+
      | GPU  GI  CI  MIG | Memory-Usage | Vol | Shared |
      | ID  ID  Dev | BAR1-Usage | SM     Unc | CE  ENC  DEC  OFA  JPG |
      |  |  | ECC |  |
      | ============================================================================= |
      | No MIG devices found |
      +-----------------------------------------------------------------------------+
      +-----------------------------------------------------------------------------+
      | Processes: |
      | GPU   GI   CI       PID   Type   Process name                   GPU Memory |
      | ID   ID                                                   Usage |
      | ============================================================================= |
      | No running processes found |
      +-----------------------------------------------------------------------------+
      Code block. nvidia-smi command - Check GPU activation status (1)
      Color mode
      $ nvidia-smi –L
      GPU 0: NVIDIA A100-SXM-80GB (UUID: GPU-c956838f-494a-92b2-6818-56eb28fe25e0)
      $ nvidia-smi –L
      GPU 0: NVIDIA A100-SXM-80GB (UUID: GPU-c956838f-494a-92b2-6818-56eb28fe25e0)
      Code block. nvidia-smi command - Check GPU activation status (2)

GPU Instance creation

If you have enabled MIG and verified its status, you can create a GPU Instance.

  1. Check the list of MIG GPU Instance profiles that can be created.
    Color mode
    $ nvidia-smi mig -i [GPU ID] -lgip
    $ nvidia-smi mig -i [GPU ID] -lgip
    Code block. nvidia-smi command - view MIG GPU Instance profile list
Color mode
$ nvidia-smi mig -i 0 -lgip
+-----------------------------------------------------------------------------+
| GPU instance profiles: |
| GPU   Name             ID    Instances   Memory     P2P    SM    DEC   ENC |
| Free/Total   GiB              CE    JPEG  OFA |
| ============================================================================= |
| 0 MIG 1g.10gb        19    7/7         9.50       No     14     0     0 |
| 1     0     0 |
+-----------------------------------------------------------------------------+
| 0 MIG 1g.10gb+me     20    1/1         9.50       No     14     0     0 |
| 1     1     1 |
+-----------------------------------------------------------------------------+
| 0 MIG 2g.20gb        14    3/3         19.50      No     28     1     0 |
| 2     0     0 |
+-----------------------------------------------------------------------------+
| 0 MIG 3g.40gb         9    2/2         39.50      No     42     2     0 |
| 3     0     0 |
+-----------------------------------------------------------------------------+
| 0 MIG 4g.40gb         5    1/1         39.50      No     56     2     0 |
| 4     0     0 |
+-----------------------------------------------------------------------------+
| 0 MIG 7g.80gb         0    1/1         79.25      No     98     0     0 |
| 7     1     1 |
+-----------------------------------------------------------------------------+
$ nvidia-smi mig -i 0 -lgip
+-----------------------------------------------------------------------------+
| GPU instance profiles: |
| GPU   Name             ID    Instances   Memory     P2P    SM    DEC   ENC |
| Free/Total   GiB              CE    JPEG  OFA |
| ============================================================================= |
| 0 MIG 1g.10gb        19    7/7         9.50       No     14     0     0 |
| 1     0     0 |
+-----------------------------------------------------------------------------+
| 0 MIG 1g.10gb+me     20    1/1         9.50       No     14     0     0 |
| 1     1     1 |
+-----------------------------------------------------------------------------+
| 0 MIG 2g.20gb        14    3/3         19.50      No     28     1     0 |
| 2     0     0 |
+-----------------------------------------------------------------------------+
| 0 MIG 3g.40gb         9    2/2         39.50      No     42     2     0 |
| 3     0     0 |
+-----------------------------------------------------------------------------+
| 0 MIG 4g.40gb         5    1/1         39.50      No     56     2     0 |
| 4     0     0 |
+-----------------------------------------------------------------------------+
| 0 MIG 7g.80gb         0    1/1         79.25      No     98     0     0 |
| 7     1     1 |
+-----------------------------------------------------------------------------+
Code block. MIG GPU Instance profile list
Reference
Refer to the NVIDIA Multi-Instance GPU User Guide for GPU Instance profiles.
  1. After creating a MIG GPU Instance, check it.
    • Create GPU Instance

      Color mode
      $ nvidia-smi mig -i [GPU ID] -cgi [Profile ID]
      $ nvidia-smi mig -i [GPU ID] -cgi [Profile ID]
      code block. nvidia-smi command - GPU Instance creation
      Color mode
      $ nvidia-smi mig -i 0 -cgi 0
      Successfully created GPU instance ID 0 on GPU 0 using profile MIG 7g.80gb (ID 0)
      $ nvidia-smi mig -i 0 -cgi 0
      Successfully created GPU instance ID 0 on GPU 0 using profile MIG 7g.80gb (ID 0)
      Code block. nvidia-smi command - Example of creating a GPU Instance

    • Check GPU Instance

      Color mode
      $ nvidia-smi mig -i [GPU ID] -lgi
      $ nvidia-smi mig -i [GPU ID] -lgi
      code block. nvidia-smi command - check GPU Instance
      Color mode
      $ nvidia-smi mig -i 0 -lgi
      +--------------------------------------------------------+
      | GPU instances: |
      | GPU   Name               Profile  Instance  Placement |
      | ID       ID      Start:Size |
      | ======================================================== |
      | 0  MIG 7g.80gb            0        0         0:8 |
      +--------------------------------------------------------+
      $ nvidia-smi mig -i 0 -lgi
      +--------------------------------------------------------+
      | GPU instances: |
      | GPU   Name               Profile  Instance  Placement |
      | ID       ID      Start:Size |
      | ======================================================== |
      | 0  MIG 7g.80gb            0        0         0:8 |
      +--------------------------------------------------------+
      Code block. nvidia-smi command - example of checking GPU Instance

Compute Instance creation

If you have created a GPU Instance, you can create a Compute Instance.

  1. Check the MIG Compute Instance profiles you can create.

    Color mode
    $ nvidia-smi mig -i [GPU ID] -gi [GPU Instance ID] -lcip
    $ nvidia-smi mig -i [GPU ID] -gi [GPU Instance ID] -lcip
    Code block. nvidia-smi command - Check MIG Compute Instance profile
    Color mode
    $ nvidia-smi mig -i 0 -gi 0 -lcip
    +---------------------------------------------------------------------------------+
    | Compute instance profiles: |
    | GPU     GPU     Name            Profile  Instances   Exclusive      Shared |
    | GPU   Instance                     ID    Free/Total     SM       DEC  ENC  OFA |
    | ID                                                       CE   JPEG |
    | ================================================================================= |
    | 0      0      MIG 1c.7g.80gb     0      7/5           14       5    0    1 |
    | 7    1 |
    +---------------------------------------------------------------------------------+
    | 0      0      MIG 2c.7g.80gb     1      3/3           28       5    0    1 |
    | 7    1 |
    +---------------------------------------------------------------------------------+
    | 0      0      MIG 3c.7g.80gb     2      2/2           42       5    0    1 |
    | 7    1 |
    +---------------------------------------------------------------------------------+
    | 0      0      MIG 4c.7g.80gb     3      1/1           56       5    0    1 |
    | 7    1 |
    +---------------------------------------------------------------------------------+
    | 0      0      MIG 7g.80gb        4*     1/1           98       5    0    1 |
    | 7    1 |
    +---------------------------------------------------------------------------------+
    $ nvidia-smi mig -i 0 -gi 0 -lcip
    +---------------------------------------------------------------------------------+
    | Compute instance profiles: |
    | GPU     GPU     Name            Profile  Instances   Exclusive      Shared |
    | GPU   Instance                     ID    Free/Total     SM       DEC  ENC  OFA |
    | ID                                                       CE   JPEG |
    | ================================================================================= |
    | 0      0      MIG 1c.7g.80gb     0      7/5           14       5    0    1 |
    | 7    1 |
    +---------------------------------------------------------------------------------+
    | 0      0      MIG 2c.7g.80gb     1      3/3           28       5    0    1 |
    | 7    1 |
    +---------------------------------------------------------------------------------+
    | 0      0      MIG 3c.7g.80gb     2      2/2           42       5    0    1 |
    | 7    1 |
    +---------------------------------------------------------------------------------+
    | 0      0      MIG 4c.7g.80gb     3      1/1           56       5    0    1 |
    | 7    1 |
    +---------------------------------------------------------------------------------+
    | 0      0      MIG 7g.80gb        4*     1/1           98       5    0    1 |
    | 7    1 |
    +---------------------------------------------------------------------------------+
    Code block. Example of MIG Compute Instance profile list

  2. Create and verify a MIG Compute Instance.

    • MIG Compute Instance creation
      Color mode
      $ nvidia-smi mig -i [GPU ID] -gi [GPU Instance ID] -cci [Compute Profile ID]
      $ nvidia-smi mig -i [GPU ID] -gi [GPU Instance ID] -cci [Compute Profile ID]
      code block. nvidia-smi command - Create MIG Compute Instance
      Color mode
      $ nvidia-smi mig -i 0 -gi 0 -cci 4
      Successfully created compute instance ID 0 on GPU instance ID 0 using profile MIG 7g.80gb(ID 4)
      $ nvidia-smi mig -i 0 -gi 0 -cci 4
      Successfully created compute instance ID 0 on GPU instance ID 0 using profile MIG 7g.80gb(ID 4)
      Code block. nvidia-smi command - Example of creating a MIG Compute Instance
    • Check MIG Compute Instance
      Color mode
      $ nvidia-smi mig -i [GPU ID] -gi [GPU Instance ID] –lci
      $ nvidia-smi mig -i [GPU ID] -gi [GPU Instance ID] –lci
      code block. nvidia-smi command - check MIG Compute Instance
      Color mode
      $ nvidia-smi mig -i 0 -gi 0 –lci
      +-----------------------------------------------------------------+
      | Compute instance profiles: |
      | GPU     GPU     Name            Profile  Instances   Placement |
      | GPU   Instance                     ID      ID        Start:Size |
      | ID |
      | ================================================================= |
      | 0      0      MIG 7g.80gb         4       0            0:7 |
      +-----------------------------------------------------------------+
      $ nvidia-smi mig -i 0 -gi 0 –lci
      +-----------------------------------------------------------------+
      | Compute instance profiles: |
      | GPU     GPU     Name            Profile  Instances   Placement |
      | GPU   Instance                     ID      ID        Start:Size |
      | ID |
      | ================================================================= |
      | 0      0      MIG 7g.80gb         4       0            0:7 |
      +-----------------------------------------------------------------+
      Code block. Example of checking MIG Compute Instance
      Color mode
      $ nvidia-smi –L
      GPU 0: NVIDIA A100-SXM-80GB (UUID: GPU-c956838f-494a-92b2-6818-56eb28fe25e0)
        MIG 7g.80gb     Device  0: (UUID: MIG-53e20040-758b-5ecb-948e-c626d03a9a32)
      $ nvidia-smi –L
      GPU 0: NVIDIA A100-SXM-80GB (UUID: GPU-c956838f-494a-92b2-6818-56eb28fe25e0)
        MIG 7g.80gb     Device  0: (UUID: MIG-53e20040-758b-5ecb-948e-c626d03a9a32)
      Code block. nvidia-smi command - Check GPU status (1)
      Color mode
      $ nvidia-smi
      Mon Sep 27 09:52:17 2021
      +-----------------------------------------------------------------------------+
      | NVIDIA-SMI 470.57.02    Driver Version: 470.57.02    CUDA Version: 11.4 |
      | -------------------------------+----------------------+---------------------- |
      | GPU  Name        Persistence-M | Bus-Id        Disp.A | Volatile Uncorr. ECC |
      | Fan  Temp  Perf  Pwr:Usage/Cap | Memory-Usage | GPU-Util  Compute M. |
      |  |  | MIG M. |
      | ===============================+======================+====================== |
      | 0  NVDIA A100-SXM...  Off | 00000000:05:00.0 Off | On |
      | N/A   32C   P0    49W / 400W | 0MiB / 81251MiB | N/A      Default |
      |  |  | Enabled |
      +-------------------------------+----------------------+----------------------+
      
      +-----------------------------------------------------------------------------+
      | MIG devices: |
      +-----------------------------------------------------------------------------+
      | GPU  GI  CI  MIG | Memory-Usage | Vol | Shared |
      | ID  ID  Dev | BAR1-Usage | SM     Unc | CE  ENC  DEC  OFA  JPG |
      |  |  | ECC |  |
      | ============================================================================= |
      | 0    0   0    0 | 0MiB / 81251MiB | 98      0 | 7   0    5    1    1 |
      |  | 1MiB / 13107... |  |  |
      +-----------------------------------------------------------------------------+
      +-----------------------------------------------------------------------------+
      | Processes: |
      | GPU   GI   CI       PID   Type   Process name                   GPU Memory |
      | ID   ID                                                   Usage |
      | ============================================================================= |
      | No running processes found |
      +-----------------------------------------------------------------------------+
      $ nvidia-smi
      Mon Sep 27 09:52:17 2021
      +-----------------------------------------------------------------------------+
      | NVIDIA-SMI 470.57.02    Driver Version: 470.57.02    CUDA Version: 11.4 |
      | -------------------------------+----------------------+---------------------- |
      | GPU  Name        Persistence-M | Bus-Id        Disp.A | Volatile Uncorr. ECC |
      | Fan  Temp  Perf  Pwr:Usage/Cap | Memory-Usage | GPU-Util  Compute M. |
      |  |  | MIG M. |
      | ===============================+======================+====================== |
      | 0  NVDIA A100-SXM...  Off | 00000000:05:00.0 Off | On |
      | N/A   32C   P0    49W / 400W | 0MiB / 81251MiB | N/A      Default |
      |  |  | Enabled |
      +-------------------------------+----------------------+----------------------+
      
      +-----------------------------------------------------------------------------+
      | MIG devices: |
      +-----------------------------------------------------------------------------+
      | GPU  GI  CI  MIG | Memory-Usage | Vol | Shared |
      | ID  ID  Dev | BAR1-Usage | SM     Unc | CE  ENC  DEC  OFA  JPG |
      |  |  | ECC |  |
      | ============================================================================= |
      | 0    0   0    0 | 0MiB / 81251MiB | 98      0 | 7   0    5    1    1 |
      |  | 1MiB / 13107... |  |  |
      +-----------------------------------------------------------------------------+
      +-----------------------------------------------------------------------------+
      | Processes: |
      | GPU   GI   CI       PID   Type   Process name                   GPU Memory |
      | ID   ID                                                   Usage |
      | ============================================================================= |
      | No running processes found |
      +-----------------------------------------------------------------------------+
      Code block. nvidia-smi command - Check GPU status (2)

Using MIG

  1. Use the MIG Instance to perform the Job.
    • Example of task execution
      Color mode
      $ docker run --gpus '"device=[GPU ID]:[MIG ID]"' -rm nvcr.io/nvidia/cuda nvidia-smi
      $ docker run --gpus '"device=[GPU ID]:[MIG ID]"' -rm nvcr.io/nvidia/cuda nvidia-smi
      Code block. Task execution example
    • You can see an example of the work performed as follows.
      Color mode
      $ docker run --gpus '"device=0:0"' -rm -it --network=host --shm-size=1g --ipc=host -v /root/.ssh/:/root/.ssh
      
      ================
      == TensorFlow ==
      ================
      
      NVIDIA Release 21.08-tf1 (build 26012104)
      TensorFlow Version 1.15.5
      
      Container image Copyright (c) 2021, NVIDIA CORPORATION. All right reserved.
      ...
      
      # Run Python process
      root@d622a93c9281:/workspace# python /workspace/nvidia-examples/cnn/resnet.py --num_iter 100
      ...
      PY 3.8.10 (default, Jun 2 2021, 10:49:15)
      [GCC 9.4.0]
      TF 1.15.5
      ...
      $ docker run --gpus '"device=0:0"' -rm -it --network=host --shm-size=1g --ipc=host -v /root/.ssh/:/root/.ssh
      
      ================
      == TensorFlow ==
      ================
      
      NVIDIA Release 21.08-tf1 (build 26012104)
      TensorFlow Version 1.15.5
      
      Container image Copyright (c) 2021, NVIDIA CORPORATION. All right reserved.
      ...
      
      # Run Python process
      root@d622a93c9281:/workspace# python /workspace/nvidia-examples/cnn/resnet.py --num_iter 100
      ...
      PY 3.8.10 (default, Jun 2 2021, 10:49:15)
      [GCC 9.4.0]
      TF 1.15.5
      ...
      Code block. Operation result
  2. Check the GPU usage. (Create JOB process)
    • When the job runs, you can see that a process is allocated to the MIG device and its utilization increases.
      Color mode
      $ nvidia-smi mig -i [GPU ID] -gi [GPU Instance ID] -lcip
      $ nvidia-smi mig -i [GPU ID] -gi [GPU Instance ID] -lcip
      code block. nvidia-smi command - check GPU utilization
    • You can check the GPU usage as shown below.
      Color mode
      +-----------------------------------------------------------------------------+
      | MIG devices: |
      +-----------------------------------------------------------------------------+
      | GPU  GI  CI  MIG | Memory-Usage | Vol | Shared |
      | ID  ID  Dev | BAR1-Usage | SM     Unc | CE  ENC  DEC  OFA  JPG |
      |  |  | ECC |  |
      | ============================================================================= |
      | 0    0   0    0 | 66562MiB / 81251MiB | 98      0 | 7   0    5    1    1 |
      |  | 5MiB / 13107... |  |  |
      +-----------------------------------------------------------------------------+
      +-----------------------------------------------------------------------------+
      | Processes: |
      | GPU   GI   CI       PID   Type   Process name                   GPU Memory |
      | ID   ID                                                   Usage |
      | ============================================================================= |
      | 0     0    0     17483      C   python                           66559MiB |
      +-----------------------------------------------------------------------------+
      +-----------------------------------------------------------------------------+
      | MIG devices: |
      +-----------------------------------------------------------------------------+
      | GPU  GI  CI  MIG | Memory-Usage | Vol | Shared |
      | ID  ID  Dev | BAR1-Usage | SM     Unc | CE  ENC  DEC  OFA  JPG |
      |  |  | ECC |  |
      | ============================================================================= |
      | 0    0   0    0 | 66562MiB / 81251MiB | 98      0 | 7   0    5    1    1 |
      |  | 5MiB / 13107... |  |  |
      +-----------------------------------------------------------------------------+
      +-----------------------------------------------------------------------------+
      | Processes: |
      | GPU   GI   CI       PID   Type   Process name                   GPU Memory |
      | ID   ID                                                   Usage |
      | ============================================================================= |
      | 0     0    0     17483      C   python                           66559MiB |
      +-----------------------------------------------------------------------------+
      Code block. Example of checking GPU utilization.

Delete and release MIG Instance

Follow these steps to delete the MIG instance and detach the MIG.

MIG release order
Delete Compute Instance → Delete GPU Instance → Disable MIG feature (deactivate)

Compute Instance Delete

  • Delete the Compute Instance.
    Color mode
    $ nvidia-smi mig -i [GPU ID] -gi [GPU Instance ID] –dci
    $ nvidia-smi mig -i [GPU ID] -gi [GPU Instance ID] -ci [Compute Instance] –dci
    $ nvidia-smi mig -i [GPU ID] -gi [GPU Instance ID] –dci
    $ nvidia-smi mig -i [GPU ID] -gi [GPU Instance ID] -ci [Compute Instance] –dci
    Code block. nvidia-smi command - Delete Compute Instance
    Color mode
    $ nvidia-smi mig -i 0 -gi 0 –lci
    +-----------------------------------------------------------------+
    | Compute instance profiles: |
    | GPU     GPU     Name            Profile  Instances   Placement |
    | GPU   Instance                     ID      ID        Start:Size |
    | ID |
    | ================================================================= |
    | 0      0      MIG 7g.80gb         4       0            0:7 |
    +-----------------------------------------------------------------+
    $ nvidia-smi mig -i 0 -gi 0 –lci
    +-----------------------------------------------------------------+
    | Compute instance profiles: |
    | GPU     GPU     Name            Profile  Instances   Placement |
    | GPU   Instance                     ID      ID        Start:Size |
    | ID |
    | ================================================================= |
    | 0      0      MIG 7g.80gb         4       0            0:7 |
    +-----------------------------------------------------------------+
    code block. Example of checking MIG Compute Instance
    Color mode
    $ nvidia-smi mig -i 0 -gi 0 –dci
    Successfully destroyed compute instance ID  0 from GPU instance ID  0
    $ nvidia-smi mig -i 0 -gi 0 –dci
    Successfully destroyed compute instance ID  0 from GPU instance ID  0
    Code block. Compute Instance deletion example
    Color mode
    $ nvidia-smi mig -i 0 -gi 0 –lci
    No compute instances found: Not found
    $ nvidia-smi mig -i 0 -gi 0 –lci
    No compute instances found: Not found
    Code block. Confirm Compute Instance deletion

Delete GPU Instance

  • Delete the GPU Instance.
    Color mode
    $ nvidia-smi mig -i [GPU ID] –dgi
    $ nvidia-smi mig -i [GPU ID] -gi [GPU Instance ID] –dgi
    $ nvidia-smi mig -i [GPU ID] –dgi
    $ nvidia-smi mig -i [GPU ID] -gi [GPU Instance ID] –dgi
    Code block. nvidia-smi command - Delete GPU Instance
    Color mode
    $ nvidia-smi mig -i 0 -lgi
    +--------------------------------------------------------+
    | GPU instances: |
    | GPU   Name               Profile  Instance  Placement |
    | ID       ID      Start:Size |
    | ======================================================== |
    | 0  MIG 7g.80gb            0        0         0:8 |
    +--------------------------------------------------------+
    $ nvidia-smi mig -i 0 -lgi
    +--------------------------------------------------------+
    | GPU instances: |
    | GPU   Name               Profile  Instance  Placement |
    | ID       ID      Start:Size |
    | ======================================================== |
    | 0  MIG 7g.80gb            0        0         0:8 |
    +--------------------------------------------------------+
    Code block. nvidia-smi command - Example of checking GPU Instance
    Color mode
    $ nvidia-smi mig -i 0 -dgi
    Successfully destroyed GPU instance ID  0 from GPU  0
    $ nvidia-smi mig -i 0 -dgi
    Successfully destroyed GPU instance ID  0 from GPU  0
    Code block. nvidia-smi command - GPU Instance deletion example
    Color mode
    $ nvidia-smi mig -i 0 -lgi
    No GPU instances found: Not found
    $ nvidia-smi mig -i 0 -lgi
    No GPU instances found: Not found
    code block. nvidia-smi command - example of deleting a GPU Instance

Disable MIG feature (deactivation)

  • After disabling MIG (Disable), reboot.
    Color mode
    $ nvidia-smi -mig 0
    Disabled MIG Mode for GPU 00000000:05:00.0
    
    All done.
    $ nvidia-smi -mig 0
    Disabled MIG Mode for GPU 00000000:05:00.0
    
    All done.
    Code block. nvidia-smi command - disable MIG
    Color mode
    $ nvidia-smi
    Mon Sep 30 05:18:28 2021
    +-----------------------------------------------------------------------------+
    | NVIDIA-SMI 470.57.02    Driver Version: 470.57.02    CUDA Version: 11.4 |
    | -------------------------------+----------------------+---------------------- |
    | GPU  Name        Persistence-M | Bus-Id        Disp.A | Volatile Uncorr. ECC |
    | Fan  Temp  Perf  Pwr:Usage/Cap | Memory-Usage | GPU-Util  Compute M. |
    |  |  | MIG M. |
    | ===============================+======================+====================== |
    | 0  NVDIA A100-SXM...  Off | 00000000:05:00.0 Off | 0 |
    | N/A   33C   P0    60W / 400W | 0MiB / 81251MiB | 0%      Default |
    |  |  | Disabled |
    +-------------------------------+----------------------+----------------------+
    +-----------------------------------------------------------------------------+
    | MIG devices: |
    +-----------------------------------------------------------------------------+
    | GPU  GI  CI  MIG | Memory-Usage | Vol | Shared |
    | ID  ID  Dev | BAR1-Usage | SM     Unc | CE  ENC  DEC  OFA  JPG |
    |  |  | ECC |  |
    | ============================================================================= |
    | No MIG devices found |
    +-----------------------------------------------------------------------------+
    +-----------------------------------------------------------------------------+
    | Processes: |
    | GPU   GI   CI       PID   Type   Process name                   GPU Memory |
    | ID   ID                                                   Usage |
    | ============================================================================= |
    | No running processes found |
    +-----------------------------------------------------------------------------+
    $ nvidia-smi
    Mon Sep 30 05:18:28 2021
    +-----------------------------------------------------------------------------+
    | NVIDIA-SMI 470.57.02    Driver Version: 470.57.02    CUDA Version: 11.4 |
    | -------------------------------+----------------------+---------------------- |
    | GPU  Name        Persistence-M | Bus-Id        Disp.A | Volatile Uncorr. ECC |
    | Fan  Temp  Perf  Pwr:Usage/Cap | Memory-Usage | GPU-Util  Compute M. |
    |  |  | MIG M. |
    | ===============================+======================+====================== |
    | 0  NVDIA A100-SXM...  Off | 00000000:05:00.0 Off | 0 |
    | N/A   33C   P0    60W / 400W | 0MiB / 81251MiB | 0%      Default |
    |  |  | Disabled |
    +-------------------------------+----------------------+----------------------+
    +-----------------------------------------------------------------------------+
    | MIG devices: |
    +-----------------------------------------------------------------------------+
    | GPU  GI  CI  MIG | Memory-Usage | Vol | Shared |
    | ID  ID  Dev | BAR1-Usage | SM     Unc | CE  ENC  DEC  OFA  JPG |
    |  |  | ECC |  |
    | ============================================================================= |
    | No MIG devices found |
    +-----------------------------------------------------------------------------+
    +-----------------------------------------------------------------------------+
    | Processes: |
    | GPU   GI   CI       PID   Type   Process name                   GPU Memory |
    | ID   ID                                                   Usage |
    | ============================================================================= |
    | No running processes found |
    +-----------------------------------------------------------------------------+
    Code block. nvidia-smi command - check GPU status

3.2.4 - Use NVSwitch on GPU Server

After creating a GPU Server, you can enable the NVSwitch feature on the GPU Server’s VM (Guest OS) and use fast GPU-to-GPU P2P communication.

Caution
Only the GPU Server (8 GPU) and Multi-node GPU Cluster of Samsung Cloud Platform are connected with NVSwitch and NVLink.

Exploring NVIDIA NVSwitch for Multi GPU

NVLink expands I/O by directly connecting multiple GPUs within a server both bidirectionally and GPU-to-GPU. Using NVSwitch, you can connect all GPUs in a server with full NVLink bandwidth.

Checking NVSwitch operation

Check the NVIDIA Fabric Manager, NVIDIA NVLink topology, and NVIDIA NVLink Status on the GPU server.

Reference
The example for checking NVSwitch operation is explained using the A100 GPU Server (g1v128a8) as a reference.

NVIDIA Fabric Manager operating status

Verify that active (running) is displayed when operating normally.

~$ systemctl status nvidia-fabricmanager
Color mode
nvidia-fabricmanager.service - NVIDIA fabric manager service
     Loaded: loaded (/lib/systemd/system/nvidia-fabricmanager.service; enabled; vendor preset: enabled)
     Active: active (running) since Mon 2026-02-02 16:23:27 KST; 32min ago
   Main PID: 2191 (nv-fabricmanage)
      Tasks: 18 (limit: 629145)
     Memory: 18.0M
        CPU: 33.461s
     CGroup: /system.slice/nvidia-fabricmanager.service
             └─2191 /usr/bin/nv-fabricmanager -c /usr/share/nvidia/nvswitch/fabricmanager.cfg
nvidia-fabricmanager.service - NVIDIA fabric manager service
     Loaded: loaded (/lib/systemd/system/nvidia-fabricmanager.service; enabled; vendor preset: enabled)
     Active: active (running) since Mon 2026-02-02 16:23:27 KST; 32min ago
   Main PID: 2191 (nv-fabricmanage)
      Tasks: 18 (limit: 629145)
     Memory: 18.0M
        CPU: 33.461s
     CGroup: /system.slice/nvidia-fabricmanager.service
             └─2191 /usr/bin/nv-fabricmanager -c /usr/share/nvidia/nvswitch/fabricmanager.cfg
Code block. Check NVIDIA Fabric Manager status

Check NVIDIA NVLink topology

Check the NVIDIA NVLink topology.

~$ nvidia-smi topo -m
Color mode
nvidia-smi topo -m
        GPU0    GPU1    GPU2    GPU3    GPU4    GPU5    GPU6    GPU7    CPU Affinity    NUMA Affinity   GPU NUMA ID
GPU0     X      NV12    NV12    NV12    NV12    NV12    NV12    NV12    0-127   0-7             N/A
GPU1    NV12     X      NV12    NV12    NV12    NV12    NV12    NV12    0-127   0-7             N/A
GPU2    NV12    NV12     X      NV12    NV12    NV12    NV12    NV12    0-127   0-7             N/A
GPU3    NV12    NV12    NV12     X      NV12    NV12    NV12    NV12    0-127   0-7             N/A
GPU4    NV12    NV12    NV12    NV12     X      NV12    NV12    NV12    0-127   0-7             N/A
GPU5    NV12    NV12    NV12    NV12    NV12     X      NV12    NV12    0-127   0-7             N/A
GPU6    NV12    NV12    NV12    NV12    NV12    NV12     X      NV12    0-127   0-7             N/A
GPU7    NV12    NV12    NV12    NV12    NV12    NV12    NV12     X      0-127   0-7             N/A

Legend:

  X    = Self
  SYS  = Connection traversing PCIe as well as the SMP interconnect between NUMA nodes (e.g., QPI/UPI)
  NODE = Connection traversing PCIe as well as the interconnect between PCIe Host Bridges within a NUMA node
  PHB  = Connection traversing PCIe as well as a PCIe Host Bridge (typically the CPU)
  PXB  = Connection traversing multiple PCIe bridges (without traversing the PCIe Host Bridge)
  PIX  = Connection traversing at most a single PCIe bridge
  NV#  = Connection traversing a bonded set of # NVLinks
nvidia-smi topo -m
        GPU0    GPU1    GPU2    GPU3    GPU4    GPU5    GPU6    GPU7    CPU Affinity    NUMA Affinity   GPU NUMA ID
GPU0     X      NV12    NV12    NV12    NV12    NV12    NV12    NV12    0-127   0-7             N/A
GPU1    NV12     X      NV12    NV12    NV12    NV12    NV12    NV12    0-127   0-7             N/A
GPU2    NV12    NV12     X      NV12    NV12    NV12    NV12    NV12    0-127   0-7             N/A
GPU3    NV12    NV12    NV12     X      NV12    NV12    NV12    NV12    0-127   0-7             N/A
GPU4    NV12    NV12    NV12    NV12     X      NV12    NV12    NV12    0-127   0-7             N/A
GPU5    NV12    NV12    NV12    NV12    NV12     X      NV12    NV12    0-127   0-7             N/A
GPU6    NV12    NV12    NV12    NV12    NV12    NV12     X      NV12    0-127   0-7             N/A
GPU7    NV12    NV12    NV12    NV12    NV12    NV12    NV12     X      0-127   0-7             N/A

Legend:

  X    = Self
  SYS  = Connection traversing PCIe as well as the SMP interconnect between NUMA nodes (e.g., QPI/UPI)
  NODE = Connection traversing PCIe as well as the interconnect between PCIe Host Bridges within a NUMA node
  PHB  = Connection traversing PCIe as well as a PCIe Host Bridge (typically the CPU)
  PXB  = Connection traversing multiple PCIe bridges (without traversing the PCIe Host Bridge)
  PIX  = Connection traversing at most a single PCIe bridge
  NV#  = Connection traversing a bonded set of # NVLinks
Code block. Check NVIDIA NVLink topology

Check NVIDIA NVLink Status

Check the NVIDIA NVLink Status.

~$ nvidia-smi topo -m
Color mode
GPU 1: NVIDIA A100-SXM4-80GB (UUID: GPU-64a2f685-bb12-c4af-105c-0726ece9c8d7)
         Link 0: 25 GB/s
         Link 1: 25 GB/s
         Link 2: 25 GB/s
         Link 3: 25 GB/s
         Link 4: 25 GB/s
         Link 5: 25 GB/s
         Link 6: 25 GB/s
         Link 7: 25 GB/s
         Link 8: 25 GB/s
         Link 9: 25 GB/s
         Link 10: 25 GB/s
         Link 11: 25 GB/s
GPU 2: NVIDIA A100-SXM4-80GB (UUID: GPU-2269851b-71cd-f6c7-50c5-ba1525cf3ce8)
         Link 0: 25 GB/s
         Link 1: 25 GB/s
         Link 2: 25 GB/s
         Link 3: 25 GB/s
         Link 4: 25 GB/s
         Link 5: 25 GB/s
         Link 6: 25 GB/s
         Link 7: 25 GB/s
         Link 8: 25 GB/s
         Link 9: 25 GB/s
         Link 10: 25 GB/s
         Link 11: 25 GB/s
GPU 3: NVIDIA A100-SXM4-80GB (UUID: GPU-4c397bbf-95fc-5c29-918a-a429cbe45a7a)
         Link 0: 25 GB/s
         Link 1: 25 GB/s
         Link 2: 25 GB/s
         Link 3: 25 GB/s
         Link 4: 25 GB/s
         Link 5: 25 GB/s
         Link 6: 25 GB/s
         Link 7: 25 GB/s
         Link 8: 25 GB/s
         Link 9: 25 GB/s
         Link 10: 25 GB/s
         Link 11: 25 GB/s
GPU 4: NVIDIA A100-SXM4-80GB (UUID: GPU-0e350204-9fb6-2cbe-538e-8f7849658eb8)
         Link 0: 25 GB/s
         Link 1: 25 GB/s
         Link 2: 25 GB/s
         Link 3: 25 GB/s
         Link 4: 25 GB/s
         Link 5: 25 GB/s
         Link 6: 25 GB/s
         Link 7: 25 GB/s
         Link 8: 25 GB/s
         Link 9: 25 GB/s
         Link 10: 25 GB/s
         Link 11: 25 GB/s
GPU 5: NVIDIA A100-SXM4-80GB (UUID: GPU-45f0c453-4760-edd4-3af9-25c5ea7473a5)
         Link 0: 25 GB/s
         Link 1: 25 GB/s
         Link 2: 25 GB/s
         Link 3: 25 GB/s
         Link 4: 25 GB/s
         Link 5: 25 GB/s
         Link 6: 25 GB/s
         Link 7: 25 GB/s
         Link 8: 25 GB/s
         Link 9: 25 GB/s
         Link 10: 25 GB/s
         Link 11: 25 GB/s
GPU 6: NVIDIA A100-SXM4-80GB (UUID: GPU-38409794-bb34-430e-3c50-90b42cb2bb72)
         Link 0: 25 GB/s
         Link 1: 25 GB/s
         Link 2: 25 GB/s
         Link 3: 25 GB/s
         Link 4: 25 GB/s
         Link 5: 25 GB/s
         Link 6: 25 GB/s
         Link 7: 25 GB/s
         Link 8: 25 GB/s
         Link 9: 25 GB/s
         Link 10: 25 GB/s
         Link 11: 25 GB/s
GPU 7: NVIDIA A100-SXM4-80GB (UUID: GPU-3fb478aa-801b-eb64-55c2-0ffc3f2ce404)
         Link 0: 25 GB/s
         Link 1: 25 GB/s
         Link 2: 25 GB/s
         Link 3: 25 GB/s
         Link 4: 25 GB/s
         Link 5: 25 GB/s
         Link 6: 25 GB/s
         Link 7: 25 GB/s
         Link 8: 25 GB/s
         Link 9: 25 GB/s
         Link 10: 25 GB/s
         Link 11: 25 GB/s
GPU 1: NVIDIA A100-SXM4-80GB (UUID: GPU-64a2f685-bb12-c4af-105c-0726ece9c8d7)
         Link 0: 25 GB/s
         Link 1: 25 GB/s
         Link 2: 25 GB/s
         Link 3: 25 GB/s
         Link 4: 25 GB/s
         Link 5: 25 GB/s
         Link 6: 25 GB/s
         Link 7: 25 GB/s
         Link 8: 25 GB/s
         Link 9: 25 GB/s
         Link 10: 25 GB/s
         Link 11: 25 GB/s
GPU 2: NVIDIA A100-SXM4-80GB (UUID: GPU-2269851b-71cd-f6c7-50c5-ba1525cf3ce8)
         Link 0: 25 GB/s
         Link 1: 25 GB/s
         Link 2: 25 GB/s
         Link 3: 25 GB/s
         Link 4: 25 GB/s
         Link 5: 25 GB/s
         Link 6: 25 GB/s
         Link 7: 25 GB/s
         Link 8: 25 GB/s
         Link 9: 25 GB/s
         Link 10: 25 GB/s
         Link 11: 25 GB/s
GPU 3: NVIDIA A100-SXM4-80GB (UUID: GPU-4c397bbf-95fc-5c29-918a-a429cbe45a7a)
         Link 0: 25 GB/s
         Link 1: 25 GB/s
         Link 2: 25 GB/s
         Link 3: 25 GB/s
         Link 4: 25 GB/s
         Link 5: 25 GB/s
         Link 6: 25 GB/s
         Link 7: 25 GB/s
         Link 8: 25 GB/s
         Link 9: 25 GB/s
         Link 10: 25 GB/s
         Link 11: 25 GB/s
GPU 4: NVIDIA A100-SXM4-80GB (UUID: GPU-0e350204-9fb6-2cbe-538e-8f7849658eb8)
         Link 0: 25 GB/s
         Link 1: 25 GB/s
         Link 2: 25 GB/s
         Link 3: 25 GB/s
         Link 4: 25 GB/s
         Link 5: 25 GB/s
         Link 6: 25 GB/s
         Link 7: 25 GB/s
         Link 8: 25 GB/s
         Link 9: 25 GB/s
         Link 10: 25 GB/s
         Link 11: 25 GB/s
GPU 5: NVIDIA A100-SXM4-80GB (UUID: GPU-45f0c453-4760-edd4-3af9-25c5ea7473a5)
         Link 0: 25 GB/s
         Link 1: 25 GB/s
         Link 2: 25 GB/s
         Link 3: 25 GB/s
         Link 4: 25 GB/s
         Link 5: 25 GB/s
         Link 6: 25 GB/s
         Link 7: 25 GB/s
         Link 8: 25 GB/s
         Link 9: 25 GB/s
         Link 10: 25 GB/s
         Link 11: 25 GB/s
GPU 6: NVIDIA A100-SXM4-80GB (UUID: GPU-38409794-bb34-430e-3c50-90b42cb2bb72)
         Link 0: 25 GB/s
         Link 1: 25 GB/s
         Link 2: 25 GB/s
         Link 3: 25 GB/s
         Link 4: 25 GB/s
         Link 5: 25 GB/s
         Link 6: 25 GB/s
         Link 7: 25 GB/s
         Link 8: 25 GB/s
         Link 9: 25 GB/s
         Link 10: 25 GB/s
         Link 11: 25 GB/s
GPU 7: NVIDIA A100-SXM4-80GB (UUID: GPU-3fb478aa-801b-eb64-55c2-0ffc3f2ce404)
         Link 0: 25 GB/s
         Link 1: 25 GB/s
         Link 2: 25 GB/s
         Link 3: 25 GB/s
         Link 4: 25 GB/s
         Link 5: 25 GB/s
         Link 6: 25 GB/s
         Link 7: 25 GB/s
         Link 8: 25 GB/s
         Link 9: 25 GB/s
         Link 10: 25 GB/s
         Link 11: 25 GB/s
Code block. Check NVIDIA NVLink status

3.2.5 - Install ServiceWatch Agent

Users can install the ServiceWatch Agent on a GPU server to collect custom metrics and logs.

Reference
Custom metric/log collection via the ServiceWatch Agent is currently available only on Samsung Cloud Platform For Enterprise. It will also be available in other offerings in the future.
Caution
Metric collection through the ServiceWatch Agent is classified as custom metrics and incurs charges, unlike the default metrics collected from each service; therefore, we recommend removing or disabling any unnecessary metric collection settings.

ServiceWatch Agent

There are two main types of agents that need to be installed on a GPU server to collect custom metrics and logs for ServiceWatch. It is the Prometheus Exporter and Open Telemetry Collector.

CategoryDetailed description
Prometheus ExporterProvide metrics of a specific application or service in a format that Prometheus can scrape
  • To collect OS metrics from servers, you can use Node Exporter for Linux servers and Windows Exporter for Windows servers, depending on the OS type.
Open Telemetry CollectorActs as a centralized collector that gathers telemetry data such as metrics and logs from distributed systems, processes (filtering, sampling, etc.) it, and exports it to various backends (e.g., Prometheus, Jaeger, Elasticsearch, etc.)
  • Exports data via the ServiceWatch Gateway so that ServiceWatch can collect metric and log data.
Table. Explanation of Prometheus Exporter and Open Telemetry Collector
Caution

If you have configured a Kubernetes Engine on a GPU server, please view the GPU metrics using the metrics provided by the Kubernetes Engine.

  • If you install the DCGM Exporter on a GPU server configured with Kubernetes Engine, it may not operate correctly.

Pre-configuration for using ServiceWatch Agent

To use the ServiceWatch Agent, please refer to ServiceWatch Agent를 위한 사전 환경 설정 and prepare the prerequisite configuration.

Install Prometheus Exporter for GPU metrics (for Ubuntu)

Install the Prometheus Exporter for collecting metrics from the GPU server according to the steps below.

Verify NVDIA Driver Installation

  • Check the installed NVDIA driver.
    Color mode
    nvidia-smi --query-gpu driver_version --format csv
    nvidia-smi --query-gpu driver_version --format csv
    Code block. NVDIA Driver version check command
    Color mode
    driver_version
    535.183.06
    
    535.183.06
    driver_version
    535.183.06
    
    535.183.06
    code block. Example of checking NVDIA Driver version.

NVSwitch Configuration and Query (NSCQ) Library Installation

Reference
The NVSwitch Configuration and Query (NSCQ) Library is required for Hopper or earlier generation GPUs.
information
The following installation commands are applicable in environments with internet access. If you are in an environment without internet access, you must download libnvdia-nscq from https://developer.download.nvidia.com/compute/cuda/repos/ and upload it.
  1. Install cuda-keyring.

    Color mode
    wget https://developer.download.nvidia.com/compute/cuda/repos/<distro>/<arch>/cuda-keyring_1.1-1_all.deb
    wget https://developer.download.nvidia.com/compute/cuda/repos/<distro>/<arch>/cuda-keyring_1.1-1_all.deb
    code block. NSCQ library download command
    Color mode
    sudo dpkg -i cuda-keyring_1.1-1_all.deb
    apt update
    sudo dpkg -i cuda-keyring_1.1-1_all.deb
    apt update
    Code block. NSCQ library installation command
    Color mode
    nvidia-smi --query-gpu driver_version --format csv
    nvidia-smi --query-gpu driver_version --format csv
    Code block. NVDIA Driver version check command
    Color mode
    driver_version
    535.183.06
    ...
    
    535.183.06
    driver_version
    535.183.06
    ...
    
    535.183.06
    Code block. NVDIA Driver version check example

  2. Install libnvidia-nscq.

    Color mode
    apt-cache policy libnvidia-nscq-535
    apt-cache policy libnvidia-nscq-535
    Code block. NSCQ library apt-cache command
    Color mode
    libnvidia-nscq-535:
      Installed: (none)
      Candidate: 535.247.01-1
      Version table:
         535.247.01-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
    ...
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
         535.216.01-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
         535.183.06-1 600  # Install the version that matches the Driver
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
         535.183.01-1 600
    ...
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
         535.54.03-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
    libnvidia-nscq-535:
      Installed: (none)
      Candidate: 535.247.01-1
      Version table:
         535.247.01-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
    ...
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
         535.216.01-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
         535.183.06-1 600  # Install the version that matches the Driver
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
         535.183.01-1 600
    ...
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
         535.54.03-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
    code block. NSCQ library apt-cache command result
    Color mode
    apt install libnvidia-nscq-535=535.183.06-1
    apt install libnvidia-nscq-535=535.183.06-1
    Code block. NSCQ library installation command

information

It must be installed with the same version as the NVDIA driver.

  • Example) driver version: 535.183.06, libnvdia-nscq version: 535.183.06-1

NVSwitch Device Monitoring API(NVSDM) Library Installation

Reference
After Blackwell, GPU architectures require the NVSDM Library to be installed. NVDIA Driver version 560 or lower does not provide the NVSDM Library.
  • Install the NVSDM library.
    Color mode
    apt-cache policy libnvsdm
    apt-cache policy libnvsdm
    code block. NVSDM library apt-cache command
    Color mode
    libnvsdm:
      Installed: (none)
      Candidate: 580.105.08-1
      Version table:
         580.105.08-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
         580.95.05-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
         580.82.07-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
         580.65.06-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
    libnvsdm:
      Installed: (none)
      Candidate: 580.105.08-1
      Version table:
         580.105.08-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
         580.95.05-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
         580.82.07-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
         580.65.06-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
    Code block. NVSDM library apt-cache command result
    Color mode
    apt install libnvsdm=580.105.08-1
    apt install libnvsdm=580.105.08-1
    Code block. Install NVSDM library

NVIDIA DCGM installation (for Ubuntu)

Install the DCGM Exporter according to the steps below.

  1. DCGM(datacenter-gpu-manager) Installation
  2. datacenter-gpu-manager-exporter Installation
  3. DCGM Service Activation and Start

DCGM(datacenter-gpu-manager) Installation

refers to a specific version of NVIDIA’s Data Center GPU Manager (DCGM) tool, which is a package for managing and monitoring NVIDIA data center GPUs. In particular, cuda12 indicates that this management tool is installed for the CUDA 12 version, and datacenter-gpu-manager-4 refers to the 4.x version of DCGM. This tool provides various features, including GPU status monitoring, diagnostics, alert system, and power/clock management.

  1. Check the CUDA version.
    Color mode
    nvidia-smi | grep CUDA
    nvidia-smi | grep CUDA
    Code block. Check CUDA version
    Color mode
    | NVIDIA-SMI 535.183.06             Driver Version: 535.183.06     CUDA Version: 12.2 |
    | NVIDIA-SMI 535.183.06             Driver Version: 535.183.06     CUDA Version: 12.2 |
    Code block. Example of CUDA version check result
    Color mode
    CUDA_VERSION=12
    CUDA_VERSION=12
    Code block. CUDA version setting command
  2. Install datacenter-gpu-manager-cuda.
    Color mode
    apt install datacenter-gpu-manager-4-cuda${CUDA_VERSION}
    apt install datacenter-gpu-manager-4-cuda${CUDA_VERSION}
    code block. datacenter-gpu-manager-cuda installation command

datacenter-gpu-manager-exporter installation

It is a tool that, based on NVIDIA Data Center GPU Manager (DCGM), collects various GPU metrics such as GPU usage, memory usage, temperature, and power consumption, and exposes them for use in monitoring systems like Prometheus.

  1. Install datacenter-gpu-manager-exporter.
    Color mode
    apt install datacenter-gpu-manager-exporter
    apt install datacenter-gpu-manager-exporter
    Code block. datacenter-gpu-manager-exporter installation command
  2. Check the DCGM Exporter configuration file.
    Color mode
    cat /usr/lib/systemd/system/nvidia-dcgm-exporter.service | grep ExecStart
    cat /usr/lib/systemd/system/nvidia-dcgm-exporter.service | grep ExecStart
    Code block. Command to check the datacenter-gpu-manager-exporter configuration file.
    Color mode
    ExecStart=/usr/bin/dcgm-exporter -f /etc/dcgm-exporter/default-counters.csv
    ExecStart=/usr/bin/dcgm-exporter -f /etc/dcgm-exporter/default-counters.csv
    Code block. Example of checking the datacenter-gpu-manager-exporter configuration file.
  3. When installing the DCGM Exporter, review the provided settings and remove # for required metrics, and add # for unnecessary metrics.
    Color mode
    vi /etc/dcgm-exporter/default-counters.csv
    ## Example ##
    
    DCGM_FI_PROF_PIPE_TENSOR_ACTIVE, gauge, Ratio of cycles the tensor (HMMA) pipe is active.
    DCGM_FI_PROF_DRAM_ACTIVE,        gauge, Ratio of cycles the device memory interface is active sending or receiving data.
    # DCGM_FI_PROF_PIPE_FP64_ACTIVE,   gauge, Ratio of cycles the fp64 pipes are active.
    # DCGM_FI_PROF_PIPE_FP32_ACTIVE,   gauge, Ratio of cycles the fp32 pipes are active.
    vi /etc/dcgm-exporter/default-counters.csv
    ## Example ##
    
    DCGM_FI_PROF_PIPE_TENSOR_ACTIVE, gauge, Ratio of cycles the tensor (HMMA) pipe is active.
    DCGM_FI_PROF_DRAM_ACTIVE,        gauge, Ratio of cycles the device memory interface is active sending or receiving data.
    # DCGM_FI_PROF_PIPE_FP64_ACTIVE,   gauge, Ratio of cycles the fp64 pipes are active.
    # DCGM_FI_PROF_PIPE_FP32_ACTIVE,   gauge, Ratio of cycles the fp32 pipes are active.
    Code block. datacenter-gpu-manager-exporter metric configuration example
Reference
For the metrics that can be collected with the GPU DCGM Exporter and how to configure them, see DCGM Exporter 지표.
Caution
Since metric collection through the ServiceWatch Agent is classified as custom metrics and incurs charges unlike the default metrics, unnecessary metric collection must be removed or disabled to prevent excessive fees.

Enable and start DCGM service

  1. Enable and start the nvdia-dcgm service.

    Color mode
    systemctl enable --now nvidia-dcgm
    systemctl enable --now nvidia-dcgm
    Code block. nvdia-dcgm service activation and start command

  2. Enable and start the nvdia-dcgm-exporter service.

    Color mode
    systemctl enable --now nvidia-dcgm-exporter
    systemctl enable --now nvidia-dcgm-exporter
    Code block. Command to enable and start the nvdia-dcgm-exporter service

Information
If you have completed the DCGM Exporter setup, you must install the OpenTelemetry Collector provided by ServiceWatch to finish configuring the ServiceWatch Agent.
For more details, refer to ServiceWatch > ServiceWatch Agent 사용하기.

Installation of Prometheus Exporter for GPU metrics (for RHEL)

Install the ServiceWatch Agent according to the steps below to collect metrics from the GPU server.

NVDIA Driver Installation Check (for RHEL)

  1. Check the installed NVDIA Driver.
    Color mode
    nvidia-smi --query-gpu driver_version --format csv
    nvidia-smi --query-gpu driver_version --format csv
    Code block. NVDIA Driver version check command
    Color mode
    driver_version
    535.183.06
    ...
    
    535.183.06
    driver_version
    535.183.06
    ...
    
    535.183.06
    Code block. Example of checking NVDIA Driver version

NVSwitch Configuration and Query (NSCQ) Library Installation (for RHEL)

Reference

NVSwitch Configuration and Query (NSCQ) Library is required for Hopper or earlier generation GPUs.

  • For RHEL, verify that libnvdia-nscq is installed and install it if necessary.
guide
The following installation commands are applicable in environments with internet access. If you are in an environment without internet access, you need to download libnvdia-nscq from https://developer.download.nvidia.com/compute/cuda/repos/ and upload it.
  1. Checking the libnvdia-nscq package.

    Color mode
    rpm -qa | grep libnvidia-nscq libnvidia-nscq-535-535.183.06-1.x86_64
    rpm -qa | grep libnvidia-nscq libnvidia-nscq-535-535.183.06-1.x86_64
    Code block. Check NSCQ library package

  2. Add the CUDA Repository to DNF.

    Color mode
    dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel8/x86_64/cuda-rhel8.repo
    dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel8/x86_64/cuda-rhel8.repo
    Code block. Add DNF Repository

  3. NVDIA Driver state reset

    Color mode
    dnf module reset nvidia-driver
    dnf module reset nvidia-driver
    Code block. Initialize the state of the NVIDIA Driver DNF module
    Color mode
    Updating Subscription Management repositories.
    Last metadata expiration check: 0:03:15 ago on Wed 19 Nov 2025 01:23:48 AM EST.
    Dependencies resolved.
    =============================================
    Package Architecture Version Repository Size
    =============================================
    Disabling module profiles:
    nvidia-driver/default
    nvidia-driver/fm
    Resetting modules:
    nvidia-driver
    
    Transaction Summary
    =============================================
    
    Is this ok [y/N]: y
    Updating Subscription Management repositories.
    Last metadata expiration check: 0:03:15 ago on Wed 19 Nov 2025 01:23:48 AM EST.
    Dependencies resolved.
    =============================================
    Package Architecture Version Repository Size
    =============================================
    Disabling module profiles:
    nvidia-driver/default
    nvidia-driver/fm
    Resetting modules:
    nvidia-driver
    
    Transaction Summary
    =============================================
    
    Is this ok [y/N]: y
    Code block. Example of the status reset result of the NVIDIA Driver DNF module.

  4. Activate the NVDIA Driver module.

    Color mode
    dnf module enable nvidia-driver:535-open
    dnf module enable nvidia-driver:535-open
    Code block. Enable NVDIA Driver module
    Color mode
    Updating Subscription Management repositories.
    Last metadata expiration check: 0:04:22 ago on Wed 19 Nov 2025 01:23:48 AM EST.
    Dependencies resolved.
    =============================================
    Package Architecture Version Repository Size
    =============================================
    Enabling module streams:
    nvidia-driver 535-open
    
    Transaction Summary
    =============================================
    
    Is this ok [y/N]: y
    Updating Subscription Management repositories.
    Last metadata expiration check: 0:04:22 ago on Wed 19 Nov 2025 01:23:48 AM EST.
    Dependencies resolved.
    =============================================
    Package Architecture Version Repository Size
    =============================================
    Enabling module streams:
    nvidia-driver 535-open
    
    Transaction Summary
    =============================================
    
    Is this ok [y/N]: y
    code block. NVDIA Driver module activation result example

  5. Check the libnvdia-nscq module list.

    Color mode
    dnf list libnvidia-nscq-535 --showduplicates
    dnf list libnvidia-nscq-535 --showduplicates
    Code block. Check libnvdia-nscq module list

  6. Install libnvdia-nscq.

    Color mode
    dnf install libnvidia-nscq-535-535.183.06-1
    dnf install libnvidia-nscq-535-535.183.06-1
    code block. libnvdia-nscq installation command

NVSwitch Device Monitoring API(NVSDM) Library Installation (for RHEL)

Reference
After Blackwell, GPU architectures require the installation of the NVSDM Library. NVDIA Driver versions 560 and below do not provide the NVSDM Library.
  1. Check the NVSDM library module list.

    Color mode
    dnf list libnvsdm --showduplicates
    dnf list libnvsdm --showduplicates
    Code block. Check NVSDM library module list
    Color mode
    libnvsdm:
      Installed: (none)
      Candidate: 580.105.08-1
      Version table:
         580.105.08-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
         580.95.05-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
         580.82.07-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
         580.65.06-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
    libnvsdm:
      Installed: (none)
      Candidate: 580.105.08-1
      Version table:
         580.105.08-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
         580.95.05-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
         580.82.07-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
         580.65.06-1 600
            600 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  Packages
    code block. Example of NVSDM library module list verification results

  2. Install libnvsdm.

    Color mode
    dnf install libnvsdm-580.105.08-1
    dnf install libnvsdm-580.105.08-1
    code block. Install NVSDM library
    Color mode
    Updating Subscription Management repositories.
    Last metadata expiration check: 0:08:18 ago on Wed 19 Nov 2025 01:05:28 AM EST.
    Dependencies resolved.
    =========================================================================
    Package Architecture Version Repository Size
    =========================================================================
    Installing:
    libnvsdm x86_64 580.105.08-1 cuda-rhel8-x86_64 675 k
    Installing dependencies:
    infiniband-diags x86_64 48.0-1.el8 rhel-8-for-x86_64-baseos-rpms 323 k
    libibumad x86_64 48.0-1.el8 rhel-8-for-x86_64-baseos-rpms 34 k
    
    Transaction Summary
    =========================================================================
    Install 3 Packages
    
    Total download size: 1.0 M
    Installed size: 3.2 M
    Is this ok [y/N]: y
    Updating Subscription Management repositories.
    Last metadata expiration check: 0:08:18 ago on Wed 19 Nov 2025 01:05:28 AM EST.
    Dependencies resolved.
    =========================================================================
    Package Architecture Version Repository Size
    =========================================================================
    Installing:
    libnvsdm x86_64 580.105.08-1 cuda-rhel8-x86_64 675 k
    Installing dependencies:
    infiniband-diags x86_64 48.0-1.el8 rhel-8-for-x86_64-baseos-rpms 323 k
    libibumad x86_64 48.0-1.el8 rhel-8-for-x86_64-baseos-rpms 34 k
    
    Transaction Summary
    =========================================================================
    Install 3 Packages
    
    Total download size: 1.0 M
    Installed size: 3.2 M
    Is this ok [y/N]: y
    Code block. Example of NVSDM library installation command output

NVIDIA DCGM installation (for RHEL)

Install Node Exporter according to the steps below.

  1. DCGM(datacenter-gpu-manager) Installation
  2. datacenter-gpu-manager-exporter installation
  3. DCGM Service Activation and Start

DCGM(datacenter-gpu-manager) Installation (for RHEL)

Refers to a specific version of NVIDIA’s Data Center GPU Manager (DCGM) tool, which is a package for managing and monitoring NVIDIA data center GPUs. In particular, cuda12 indicates that this management tool is installed for the CUDA 12 version, and datacenter-gpu-manager-4 refers to the 4.x version of DCGM. This tool provides various features, including GPU status monitoring, diagnostics, alert system, and power/clock management.

  1. Add the CUDA Repository to DNF.
    Color mode
    dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel8/x86_64/cuda-rhel8.repo
    dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel8/x86_64/cuda-rhel8.repo
    code block. Add DNF Repository
  2. Check the CUDA version.
    Color mode
    nvidia-smi | grep CUDA
    nvidia-smi | grep CUDA
    Code block. Check CUDA version
    Color mode
    | NVIDIA-SMI 535.183.06             Driver Version: 535.183.06     CUDA Version: 12.2 |
    | NVIDIA-SMI 535.183.06             Driver Version: 535.183.06     CUDA Version: 12.2 |
    Code block. Example of CUDA version check result
    Color mode
    CUDA_VERSION=12
    CUDA_VERSION=12
    Code block. CUDA version setting command
  3. Check the list of datacenter-gpu-manager-cuda modules.
    Color mode
    dnf list datacenter-gpu-manager-4-cuda${CUDA_VERSION} --showduplicates
    dnf list datacenter-gpu-manager-4-cuda${CUDA_VERSION} --showduplicates
    Code block. Check the datacenter-gpu-manager-cuda module list.
    Color mode
    Updating Subscription Management repositories.
    Unable to read consumer identity
    
    This system is not registered with an entitlement server. You can use subscription-manager to register.
    
    Last metadata expiration check: 0:00:34 ago on Wed 19 Nov 2025 12:26:56 AM EST.
    Available Packages
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.0.0-1    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.1.0-1    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.1.1-1    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.2.0-1    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.2.2-1    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.2.3-1    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.2.3-2    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.3.0-1    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.3.1-1    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.4.0-1    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.4.1-1    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.4.2-1    cuda-rhel8-x86_64
    Updating Subscription Management repositories.
    Unable to read consumer identity
    
    This system is not registered with an entitlement server. You can use subscription-manager to register.
    
    Last metadata expiration check: 0:00:34 ago on Wed 19 Nov 2025 12:26:56 AM EST.
    Available Packages
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.0.0-1    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.1.0-1    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.1.1-1    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.2.0-1    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.2.2-1    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.2.3-1    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.2.3-2    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.3.0-1    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.3.1-1    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.4.0-1    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.4.1-1    cuda-rhel8-x86_64
    datacenter-gpu-manager-4-cuda12.x86_64   1:4.4.2-1    cuda-rhel8-x86_64
    code block. Example of checking the module list of datacenter-gpu-manager-cuda
  4. Install datacenter-gpu-manager-cuda.
    Color mode
    dnf install datacenter-gpu-manager-4-cuda${CUDA_VERSION}
    dnf install datacenter-gpu-manager-4-cuda${CUDA_VERSION}
    code block. Install datacenter-gpu-manager-cuda
    Color mode
    Updating Subscription Management repositories.
    Unable to read consumer identity
    
    This system is not registered with an entitlement server. You can use subscription-manager to register.
    
    Last metadata expiration check: 0:07:12 ago on Wed 19 Nov 2025 12:26:56 AM EST.
    Dependencies resolved.
    ===================================================================================================
     Package                                       Architecture   Version     Repository          Size
    ===================================================================================================
    Installing:
     datacenter-gpu-manager-4-cuda12               x86_64         1:4.4.2-1   cuda-rhel8-x86_64   554 M
    Installing dependencies:
     datacenter-gpu-manager-4-core                 x86_64         1:4.4.2-1   cuda-rhel8-x86_64   9.9 M
    Installing weak dependencies:
     datacenter-gpu-manager-4-proprietary          x86_64         1:4.4.2-1   cuda-rhel8-x86_64   5.3 M
     datacenter-gpu-manager-4-proprietary-cuda12   x86_64         1:4.4.2-1   cuda-rhel8-x86_64   289 M
    
    Transaction Summary
    ====================================================================================================
    Install  4 Packages
    ...
    Is this ok [y/N]: y
    Updating Subscription Management repositories.
    Unable to read consumer identity
    
    This system is not registered with an entitlement server. You can use subscription-manager to register.
    
    Last metadata expiration check: 0:07:12 ago on Wed 19 Nov 2025 12:26:56 AM EST.
    Dependencies resolved.
    ===================================================================================================
     Package                                       Architecture   Version     Repository          Size
    ===================================================================================================
    Installing:
     datacenter-gpu-manager-4-cuda12               x86_64         1:4.4.2-1   cuda-rhel8-x86_64   554 M
    Installing dependencies:
     datacenter-gpu-manager-4-core                 x86_64         1:4.4.2-1   cuda-rhel8-x86_64   9.9 M
    Installing weak dependencies:
     datacenter-gpu-manager-4-proprietary          x86_64         1:4.4.2-1   cuda-rhel8-x86_64   5.3 M
     datacenter-gpu-manager-4-proprietary-cuda12   x86_64         1:4.4.2-1   cuda-rhel8-x86_64   289 M
    
    Transaction Summary
    ====================================================================================================
    Install  4 Packages
    ...
    Is this ok [y/N]: y
    Code block. Example of datacenter-gpu-manager-cuda installation result

datacenter-gpu-manager-exporter installation (for RHEL)

It is a tool that, based on NVIDIA Data Center GPU Manager (DCGM), collects various GPU metrics such as GPU usage, memory usage, temperature, and power consumption, and exposes them for use in monitoring systems like Prometheus.

  1. Add the CUDA Repository to DNF. 1. (If you have already performed this command, proceed to the next step.)

    Color mode
    dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel8/x86_64/cuda-rhel8.repo
    dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel8/x86_64/cuda-rhel8.repo
    code block. Add DNF Repository

  2. Check the CUDA version. 2. (If you have already performed this command, proceed to the next step.)

    Color mode
    nvidia-smi | grep CUDA
    nvidia-smi | grep CUDA
    Code block. Check CUDA version
    Color mode
    | NVIDIA-SMI 535.183.06             Driver Version: 535.183.06     CUDA Version: 12.2 |
    | NVIDIA-SMI 535.183.06             Driver Version: 535.183.06     CUDA Version: 12.2 |
    Code block. Example of CUDA version check result
    Color mode
    CUDA_VERSION=12
    CUDA_VERSION=12
    Code block. CUDA version setting command

  3. Check the list of datacenter-gpu-manager-exporter modules.

    Color mode
    dnf list datacenter-gpu-manager-exporter --showduplicates
    dnf list datacenter-gpu-manager-exporter --showduplicates
    Code block. Check the list of datacenter-gpu-manager-exporter modules
    Color mode
    Updating Subscription Management repositories.
    Unable to read consumer identity
    
    This system is not registered with an entitlement server. You can use subscription-manager to register.
    
    Last metadata expiration check: 0:02:11 ago on Wed 19 Nov 2025 12:26:56 AM EST.
    Available Packages
    datacenter-gpu-manager-exporter.x86_64   4.0.1-1   cuda-rhel8-x86_64
    datacenter-gpu-manager-exporter.x86_64   4.1.0-1   cuda-rhel8-x86_64
    datacenter-gpu-manager-exporter.x86_64   4.1.1-1   cuda-rhel8-x86_64
    datacenter-gpu-manager-exporter.x86_64   4.1.3-1   cuda-rhel8-x86_64
    datacenter-gpu-manager-exporter.x86_64   4.5.0-1   cuda-rhel8-x86_64
    datacenter-gpu-manager-exporter.x86_64   4.5.1-1   cuda-rhel8-x86_64
    datacenter-gpu-manager-exporter.x86_64   4.5.2-1   cuda-rhel8-x86_64
    datacenter-gpu-manager-exporter.x86_64   4.6.0-1   cuda-rhel8-x86_64
    Updating Subscription Management repositories.
    Unable to read consumer identity
    
    This system is not registered with an entitlement server. You can use subscription-manager to register.
    
    Last metadata expiration check: 0:02:11 ago on Wed 19 Nov 2025 12:26:56 AM EST.
    Available Packages
    datacenter-gpu-manager-exporter.x86_64   4.0.1-1   cuda-rhel8-x86_64
    datacenter-gpu-manager-exporter.x86_64   4.1.0-1   cuda-rhel8-x86_64
    datacenter-gpu-manager-exporter.x86_64   4.1.1-1   cuda-rhel8-x86_64
    datacenter-gpu-manager-exporter.x86_64   4.1.3-1   cuda-rhel8-x86_64
    datacenter-gpu-manager-exporter.x86_64   4.5.0-1   cuda-rhel8-x86_64
    datacenter-gpu-manager-exporter.x86_64   4.5.1-1   cuda-rhel8-x86_64
    datacenter-gpu-manager-exporter.x86_64   4.5.2-1   cuda-rhel8-x86_64
    datacenter-gpu-manager-exporter.x86_64   4.6.0-1   cuda-rhel8-x86_64
    code block. Example of checking the module list of datacenter-gpu-manager-exporter

  4. Install datacenter-gpu-manager-cuda. dcgm-exporter 4.5.X requires glibc 2.34 or newer, but since RHEL9 provides glibc 2.34, we specify the version as 4.1.3-1 for installation.

    Color mode
    dnf install datacenter-gpu-manager-exporter-4.1.3-1
    dnf install datacenter-gpu-manager-exporter-4.1.3-1
    code block. datacenter-gpu-manager-cuda installation
    Color mode
    Updating Subscription Management repositories.
    Unable to read consumer identity
    
    This system is not registered with an entitlement server. You can use subscription-manager to register.
    
    Last metadata expiration check: 0:07:12 ago on Wed 19 Nov 2025 12:26:56 AM EST.
    Dependencies resolved.
    ====================================================================================================
     Package                                       Architecture   Version     Repository          Size
    ====================================================================================================
    Installing:
     datacenter-gpu-manager-exporter               x86_64         4.1.3-1     cuda-rhel8-x86_64   26 M
    
    
    Is this ok [y/N]: y
    Updating Subscription Management repositories.
    Unable to read consumer identity
    
    This system is not registered with an entitlement server. You can use subscription-manager to register.
    
    Last metadata expiration check: 0:07:12 ago on Wed 19 Nov 2025 12:26:56 AM EST.
    Dependencies resolved.
    ====================================================================================================
     Package                                       Architecture   Version     Repository          Size
    ====================================================================================================
    Installing:
     datacenter-gpu-manager-exporter               x86_64         4.1.3-1     cuda-rhel8-x86_64   26 M
    
    
    Is this ok [y/N]: y
    Code block. Example of datacenter-gpu-manager-cuda installation result
    Color mode
    cat /usr/lib/systemd/system/nvidia-dcgm-exporter.service | grep ExecStart
    cat /usr/lib/systemd/system/nvidia-dcgm-exporter.service | grep ExecStart
    code block. datacenter-gpu-manager-exporter configuration file
    Color mode
    ExecStart=/usr/bin/dcgm-exporter -f /etc/dcgm-exporter/default-counters.csv
    ExecStart=/usr/bin/dcgm-exporter -f /etc/dcgm-exporter/default-counters.csv
    code block. Example of checking the datacenter-gpu-manager-exporter configuration file.

  5. When installing the DCGM Exporter, review the provided settings and remove # for required metrics, and add # for unnecessary metrics.

    Color mode
    vi /etc/dcgm-exporter/default-counters.csv
    ## Example ##
    
    DCGM_FI_PROF_PIPE_TENSOR_ACTIVE, gauge, Ratio of cycles the tensor (HMMA) pipe is active.
    DCGM_FI_PROF_DRAM_ACTIVE,        gauge, Ratio of cycles the device memory interface is active sending or receiving data.
    # DCGM_FI_PROF_PIPE_FP64_ACTIVE,   gauge, Ratio of cycles the fp64 pipes are active.
    # DCGM_FI_PROF_PIPE_FP32_ACTIVE,   gauge, Ratio of cycles the fp32 pipes are active.
    vi /etc/dcgm-exporter/default-counters.csv
    ## Example ##
    
    DCGM_FI_PROF_PIPE_TENSOR_ACTIVE, gauge, Ratio of cycles the tensor (HMMA) pipe is active.
    DCGM_FI_PROF_DRAM_ACTIVE,        gauge, Ratio of cycles the device memory interface is active sending or receiving data.
    # DCGM_FI_PROF_PIPE_FP64_ACTIVE,   gauge, Ratio of cycles the fp64 pipes are active.
    # DCGM_FI_PROF_PIPE_FP32_ACTIVE,   gauge, Ratio of cycles the fp32 pipes are active.
    Code block. datacenter-gpu-manager-exporter metric configuration example

Reference
For the metrics that can be collected with the GPU DCGM Exporter and how to configure them, see DCGM Exporter 지표.
Caution
Since metric collection through the ServiceWatch Agent is classified as custom metrics and incurs charges unlike the default metrics, unnecessary metric collection must be removed or disabled to prevent excessive fees.

Enable and start DCGM service (for RHEL)

  1. Enable and start the nvdia-dcgm service.

    Color mode
    systemctl enable --now nvidia-dcgm
    systemctl enable --now nvidia-dcgm
    code block. nvdia-dcgm service enable and start command

  2. Enable and start the nvdia-dcgm-exporter service.

    Color mode
    systemctl enable --now nvidia-dcgm-exporter
    systemctl enable --now nvidia-dcgm-exporter
    code block. nvdia-dcgm-exporter service enable and start command

Info
If you have completed the DCGM Exporter setup, you must install the OpenTelemetry Collector provided by ServiceWatch to finish configuring the ServiceWatch Agent.
For more details, refer to ServiceWatch > ServiceWatch Agent 사용하기.

DCGM Exporter metrics

DCGM Exporter Key Metrics

Among the metrics provided by the DCGM Exporter, the main GPU metrics are listed below.

CategoryDCGM FieldPrometheus Metric TypeSummary
ClocksDCGM_FI_DEV_SM_CLOCKgaugeSM clock frequency (in MHz)
ClocksDCGM_FI_DEV_MEM_CLOCKgaugeMemory clock frequency (in MHz)
TemperatureDCGM_FI_DEV_GPU_TEMPgaugeGPU temperature (in C)
PowerDCGM_FI_DEV_POWER_USAGEgaugePower draw (in W)
UtilizationDCGM_FI_DEV_GPU_UTILgaugeGPU utilization (in %)
UtilizationDCGM_FI_DEV_MEM_COPY_UTILgaugeMemory utilization (in %)
Memory UsageDCGM_FI_DEV_FB_FREEgaugeFrame buffer memory free (in MiB)
Memory UsageDCGM_FI_DEV_FB_USEDgaugeFrame buffer memory used (in MiB)
NvlinkDCGM_FI_DEV_NVLINK_BANDWIDTH_TOTAL(8 GPU only)counterTotal number of NVLink bandwidth counters for all lanes
Table. Main GPU metrics provided by the DCGM Exporter

DCGM Exporter metric collection configuration

Refer to the default metrics configured for DCGM Exporter at DCGM Exporter > 기본 지표.

  • In addition to the default settings, remove # from default-counters.csv for the additional metrics to be configured.
  • For default metrics you do not wish to collect, add # or delete the corresponding item.
Color mode
# Format
# If line starts with a '#' it is considered a comment
# DCGM FIELD, Prometheus metric type, help message

# Clocks
DCGM_FI_DEV_SM_CLOCK, gauge, SM clock frequency (in MHz).
DCGM_FI_DEV_MEM_CLOCK, gauge, Memory clock frequency (in MHz).

# Temperature
DCGM_FI_DEV_MEMORY_TEMP, gauge, Memory temperature (in C).
DCGM_FI_DEV_GPU_TEMP,    gauge, GPU temperature (in C).

# Power
DCGM_FI_DEV_POWER_USAGE,              gauge, Power draw (in W).
DCGM_FI_DEV_TOTAL_ENERGY_CONSUMPTION, counter, Total energy consumption since boot (in mJ).

# PCIE
# DCGM_FI_PROF_PCIE_TX_BYTES,  counter, Total number of bytes transmitted through PCIe TX via NVML.
# DCGM_FI_PROF_PCIE_RX_BYTES,  counter, Total number of bytes received through PCIe RX via NVML.
...
# Format
# If line starts with a '#' it is considered a comment
# DCGM FIELD, Prometheus metric type, help message

# Clocks
DCGM_FI_DEV_SM_CLOCK, gauge, SM clock frequency (in MHz).
DCGM_FI_DEV_MEM_CLOCK, gauge, Memory clock frequency (in MHz).

# Temperature
DCGM_FI_DEV_MEMORY_TEMP, gauge, Memory temperature (in C).
DCGM_FI_DEV_GPU_TEMP,    gauge, GPU temperature (in C).

# Power
DCGM_FI_DEV_POWER_USAGE,              gauge, Power draw (in W).
DCGM_FI_DEV_TOTAL_ENERGY_CONSUMPTION, counter, Total energy consumption since boot (in mJ).

# PCIE
# DCGM_FI_PROF_PCIE_TX_BYTES,  counter, Total number of bytes transmitted through PCIe TX via NVML.
# DCGM_FI_PROF_PCIE_RX_BYTES,  counter, Total number of bytes received through PCIe RX via NVML.
...
Code block. default-counters.csv configuration example

3.3 - API Reference

API Reference

3.4 - CLI Reference

CLI Reference

3.5 - Release Note

GPU Server

2026.07.16
FEATURE Add new server type and other feature changes
  • We provide NPU server types to diversify AI accelerators.
    • You can select and use an OS image for the NPU.
  • The n1 server type and the A100 HPE server type have been added.
  • Even if the Backup service is connected, you can delete the server.
    • When a server is deleted, the schedule of the associated backup is removed, but the backup itself remains intact and can be restored and utilized.

2026.03.19
FEATURE Add Kubernetes image and SSD_Provisioned disk type
  • Add Kubernetes image
    • When creating a GPU server, you can select the Kubernetes image (Ubuntu).
  • An SSD volume with configurable IOPS and Throughput has been added.
    • When creating Block Storage, you can select the SSD_Provisioned disk type.
    • You can set the maximum IOPS and Throughput.
2025.10.23
FEATURE Add new features and provide ServiceWatch service integration functionality
  • ServiceWatch service integration provision
    • You can monitor data through the ServiceWatch service.
  • When creating a GPU Server, you can select a RHEL image.
  • Keypair management feature has been added.
    • You can create a keypair for use or retrieve a public key and apply it.
2025.07.01
FEATURE Add GPU Server feature, change Image sharing method, and add GPU Server usage guide
  • Add GPU Server feature
    • IP, Public NAT IP, Private NAT IP configuration feature has been added.
    • An LLM Endpoint is provided for LLM usage.
  • The method for sharing images between accounts has changed.
    • You can create a new Image for sharing.
  • Add GPU Server usage guide
2025.04.28
FEATURE Add OS image
  • GPU Server RHEL OS and GPU driver versions have been added.
2025.02.27
FEATURE Common feature change
  • Add GPU Server feature
    • NAT configuration functionality has been added to the GPU Server.
  • Samsung Cloud Platform Common Feature Change
    • Account, IAM, and Service Home, tags, etc., have been updated to reflect common CX changes.
2024.10.01
NEW GPU Server Service Official Version Release
  • We have officially launched the GPU Server service.
  • We have launched a virtualization computing service that lets you allocate and use infrastructure resources such as CPU, GPU, and memory provided by the server as needed, without having to purchase them individually.

4 - Bare Metal Server

4.1 - Overview

Service Overview

Bare Metal Server does not use virtualization technology and is a high-performance cloud computing service that can allocate and use physically separated computing resources such as CPU and memory exclusively. It is not affected by other cloud users, allowing you to reliably operate performance-sensitive services.

Features

  • Easy and convenient computing environment setup: Through the web-based Console, you can easily handle everything from Bare Metal Server provisioning to resource management and cost management. You can receive a server with standard specs (CPU, Memory, Disk) allocated exclusively and use it immediately.

  • Providing High-Performance Computing Environment: We provide servers suitable for workloads that require large capacity and high performance—such as real-time (Real-Time) systems, HPC (High Performance Computing), and servers that demand excessive I/O usage—in a physically isolated environment.

  • Efficient Service Delivery: Ensure performance and stability through optimal server selection and in-house testing. Customers can choose the optimal resources that fit their service environment from the various specifications of Bare Metal Servers offered by Samsung Cloud Platform.

Service Architecture Diagram

Diagram
Figure. Bare Metal Server Diagram

Provided features

Bare Metal Server provides the following features.

  • Auto Provisioning (Auto Provisioning) and Management: Through the web-based Console, you can easily provision Bare Metal Servers, manage resources, and control costs.
  • Providing various types of server types and OS images: Provides CPU, Memory, and Disk resources of standard server types, and offers a variety of standard OS images.
  • Storage Connection: Provides additional attached storage beyond the OS disk. You can connect and use Block Storage, File Storage, and Object Storage.
  • Network Connection: You can connect the standard subnet/IP settings of a Bare Metal Server and a Public NAT IP. Provides a local subnet connection for inter-server communication. This can be modified on the detail page.
  • Monitoring: You can view monitoring information such as CPU, Memory, and Disk, which are computing resources, through Cloud Monitoring. To use the Cloud Monitoring service for Bare Metal Server, you must install the Agent. Please be sure to install the Agent for stable Bare Metal Server service usage. For more details, see Bare Metal Server Monitoring Metrics.
  • Backup and Recovery: You can back up and restore the Bare Metal Server’s filesystem using the Backup service.
  • Efficient Cost Management: You can easily create or terminate servers as needed, and because billing is based on actual usage time, you can use it cost‑effectively in various unpredictable situations.
  • Local disk partition creation You can create and use up to 10 local disk partitions.
  • Terraform Provisioning: Provides an IaC environment using Terraform.

Components

Bare Metal Server provides various standard OS images and standard server types. Users can select and use them according to the scale of the service they want to configure.

OS Image provided version

The OS images supported by Bare Metal Server are as follows

OS Image versionEoS Date
Oracle Linux 9.62032-06-30
RHEL 8.102029-05-31
RHEL 9.42026-04-30
RHEL 9.62027-05-31
Rocky Linux 8.102029-05-31
Rocky Linux 9.62025-11-30
Ubuntu 22.042027-06-30
Ubuntu 24.042029-06-30
Windows 20192029-01-09
Windows 20222031-10-14
Table. Bare Metal Server provided OS Image version
Reference
  • Oracle Linux, RHEL, and Rocky Linux only provide even-numbered minor versions. Check the OS Image’s EOS (End of Support)/EOL (End of Life) Date.
  • To ensure stable OS operation, apply new or additional individual packages yourself.

Server type

The server types supported by Bare Metal Server are as follows. For detailed information about the server types supported by Bare Metal Server, see Bare Metal Server Server Types.

s3v16m64_metal
CategoryexampleDetailed description
Server generations3Provided server categories and generations
  • s3: s means the standard, commonly used standard specification (vCPU, Memory) configuration, and 1 denotes the generation
CPU vCorev16Number of vCores
  • v16: Allocated vCores are twice the number of physical cores
    • 16 vCores correspond to 8 physical cores
    • Provided with Hyperthreading enabled by default, which can be disabled when creating a service
Memorym64Memory capacity
  • m64: 64GB Memory
Table. Bare Metal Server server type

Preceding Service

This is a list of services that must be pre-configured before creating the service. Please refer to the guide provided for each service and prepare in advance.

Service CategoryserviceDetailed description
NetworkingVPCA service that provides an isolated virtual network in a cloud environment
Table. Bare Metal Server Pre-service

4.1.1 - Server type

Bare Metal Server provides server types tailored to the intended use. Server types consist of various combinations such as CPU and Memory, and the server used for a Bare Metal Server is determined by the server type selected when creating the Bare Metal Server. Please select the server type based on the specifications of the application you want to run on the Bare Metal Server.

안내
  • Support for the server type varies depending on the OS image provided by the Bare Metal Server. * When creating a service, refer to OS Image별 지원 서버 타입 and select the OS Image and server type.
  • s3/h3, s2/h2 server types are no longer accepting new service applications. * There is no impact on the services currently in use.

Bare Metal Server server type

The server types supported by Bare Metal Server are as follows.

s3v16m64_metal
Category
exampleDetailed description
Server generations3Provided server categories and generations
  • s3
    • smeans standard specification
    • 3means generation
  • h3
    • hmeans high-capacity server specification
    • 3means generation
CPU vCorev16Number of vCores
  • v16: Allocated vCores are twice the number of physical cores
    • 16 vCores correspond to 8 physical cores
    • Provided with Hyper-Threading enabled by default, which can be disabled when creating a service
Memorym64메모리 용량
  • m64: 64GB Memory
표. Bare Metal Server 서버 타입 형식

s4/h4 server type

The Bare Metal Server s4 server type is offered with standard specifications (vCPU, Memory) and, because it receives physically isolated resources, it is suitable for high-performance applications. Additionally, the Bare Metal Server h4 server type is offered with high-capacity specifications and is suitable for high-performance applications for large-scale data processing.

  • Supports a total of five vCPU options (16, 32, 64, 96, 128 vCore)
    • Intel 6th Generation (Granite Rapids) Processor
  • Supports up to 64 physical cores, 128 vCPUs, and 2,048 GB of memory.
  • Provides two internal disks up to 1.92 TB (for OS use)
서버 타입물리 CPUvCPUMemoryCPU Type소켓수Internal Disk (OS)
s4v16m64_metal8 Core16 vCore64 GBIntel Xeon 6507P with a maximum frequency of 4.3 GHz1480 GB * 2 EA
s4v16m128_metal8 Core16 vCore128 GBIntel Xeon 6507P with a maximum of 4.3 GHz1480 GB * 2 EA
s4v16m256_metal8 Core16 vCore256 GBIntel Xeon 6507P up to 4.3 GHz1480 GB * 2 EA
h4v32m128_metal16 Core32 vCore128 GBIntel Xeon 6517P up to 4.0 GHz1960 GB * 2 EA
h4v32m256_metal16 Core32 vCore256 GBIntel Xeon 6517P up to 4.0 GHz1960 B * 2 EA
h4v32m512_metal16 Core32 vCore512 GBIntel Xeon 6517P with a maximum of 4.0 GHz1960 GB * 2 EA
h4v64m256_metal32 Core64 vCore256 GBIntel Xeon 6737P up to 4.0 GHz11.92 TB * 2 EA
h4v64m512_metal32 Core64 vCore512 GBIntel Xeon 6737P up to 4.0 GHz11.92 TB * 2 EA
h4v64m1024_metal32 Core64 vCore1,024 GB최대 4.0 GHz의 Intel Xeon 6737P11.92 TB * 2 EA
h4v96m512_metal48 Core96 vCore512 GBIntel Xeon 6520P with a maximum of 3.4 GHz21.92 TB * 2 EA
h4v96m768_metal48 Core96 vCore768 GBIntel Xeon 6520P with a maximum of 3.4 GHz21.92 TB * 2 EA
h4v96m2048_metal48 Core96 vCore2,048 GBIntel Xeon 6520P with a maximum of 3.4 GHz21.92 TB * 2 EA
h4v128m512_metal64 Core128 vCore512 GBIntel Xeon 6737P up to 4.0 GHz21.92 TB * 2 EA
h4v128m1024_metal64 Core128 vCore1,024 GBIntel Xeon 6737P up to 4.0 GHz21.92 TB * 2 EA
h4v128m2048_metal64 Core128 vCore2,048 GBIntel Xeon 6737P with a maximum of 4.0 GHz21.92 TB * 2 EA
표. Bare Metal Server 서버 타입 사양 > s4/h4 서버 타입

s3/h3 server type

안내
s3/h3 서버 타입은 신규 신청 서비스가 종료되었습니다. 기존에 사용 중인 서비스에는 영향이 없습니다.

The Bare Metal Server s3 server type is offered with standard specifications (vCPU, Memory) and, because it receives physically isolated resources, it is suitable for high-performance applications. Additionally, the Bare Metal Server h3 server type is offered with high-capacity specifications and is suitable for high-performance applications for large-scale data processing.

  • Supports a total of five vCPU options (16, 32, 64, 96, 128 vCore)
    • Intel 4th Generation (Sapphire Rapids) Processor
  • Supports up to 64 physical cores, 128 vCPUs, and 2,048 GB of memory.
  • Provides two internal disks up to 1.92 TB (for OS use)
Server typePhysical CPUvCPUMemoryCPU Typesocket countInternal Disk (OS)
s3v16m64_metal8 Core16 vCore64 GBIntel Xeon Gold 6434 with a maximum of 4.1 GHz1480 GB * 2 EA
s3v16m128_metal8 Core16 vCore128 GBIntel Xeon Gold 6434 with a maximum of 4.1 GHz1480 GB * 2 EA
s3v16m256_metal8 Core16 vCore256 GBIntel Xeon Gold 6434 with a maximum of 4.1 GHz1480 GB * 2 EA
h3v32m128_metal16 Core32 vCore128 GBIntel Xeon Gold 6444Y up to 4.0 GHz1960 GB * 2 EA
h3v32m256_metal16 Core32 vCore256 GBIntel Xeon Gold 6444Y up to 4.0 GHz1960 GB * 2 EA
h3v32m512_metal16 Core32 vCore512 GBIntel Xeon Gold 6444Y up to 4.0 GHz1960 GB * 2 EA
h3v64m256_metal32 Core64 vCore256 GBIntel Xeon Gold 6448H up to 3.2 GHz11.92 TB * 2 EA
h3v64m512_metal32 Core64 vCore512 GBIntel Xeon Gold 6448H with a maximum of 3.2 GHz11.92 TB * 2 EA
h3v64m1024_metal32 Core64 vCore1,024 GBIntel Xeon Gold 6448H up to 3.2 GHz11.92 TB * 2 EA
h3v96m384_metal48 Core96 vCore384 GBIntel Xeon Gold 6442Y up to 3.3 GHz21.92 TB * 2 EA
h3v96m768_metal48 Core96 vCore768 GBIntel Xeon Gold 6442Y up to 3.3 GHz21.92 TB * 2 EA
h3v96m1536_metal48 Core96 vCore1,536 GBIntel Xeon Gold 6442Y up to 3.3 GHz21.92 TB * 2 EA
h3v128m512_metal64 Core128 vCore512 GBIntel Xeon Gold 6448H up to 3.2 GHz21.92 TB * 2 EA
h3v128m1024_metal64 Core128 vCore1,024 GBIntel Xeon Gold 6448H up to 3.2 GHz21.92 TB * 2 EA
h3v128m2048_metal64 Core128 vCore2,048 GBIntel Xeon Gold 6448H up to 3.2 GHz21.92 TB * 2 EA
표. Bare Metal Server 서버 타입 사양 > s3/h3 서버 타입

s2/h2 server type

안내
s2/h2 서버 타입은 신규 신청 서비스가 종료되었습니다. 기존에 사용 중인 서비스에는 영향이 없습니다.

The Bare Metal Server s2 server type is offered with standard specifications (vCPU, Memory) and, because it receives physically isolated resources, it is suitable for high-performance applications. Additionally, the Bare Metal Server h2 server type is offered with high-capacity specifications and is suitable for high-performance applications that require large-scale data processing.

  • Supports a total of five vCPU options (16, 24, 32, 72, 96 vCore)
    • Intel 3rd Generation (Ice Lake) Processor
  • Supports up to 48 physical cores, 96 vCPUs, and 1,024 GB of memory
  • Provides two internal disks up to 1.92 TB (for OS use)
Server typePhysical CPUvCPUMemoryCPU Typesocket countInternal Disk (OS)
s2v16m64_metal8 Core16 vCore64 GBIntel Xeon Gold 6334 up to 3.6 GHz1480 GB * 2 EA
s2v16m128_metal8 Core16 vCore128 GBIntel Xeon Gold 6334 up to 3.6 GHz1480 GB * 2 EA
s2v16m256_metal8 Core16 vCore256 GBIntel Xeon Gold 6334 up to 3.6 GHz1480 GB * 2 EA
h2v24m96_metal12 Core24 vCore96 GBIntel Xeon Gold 5317 with a maximum of 3.4 GHz1960 GB * 2 EA
h2v24m192_metal12 Core24 vCore192 GBIntel Xeon Gold 5317 up to 3.4 GHz1960 GB * 2 EA
h2v24m384_metal12 Core24 vCore384 GBIntel Xeon Gold 5317 up to 3.4 GHz1960 GB * 2 EA
h2v32m128_metal16 Core32 vCore128 GBIntel Xeon Gold 6346 up to 3.6 GHz1960 GB * 2 EA
h2v32m256_metal16 Core32 vCore256 GBIntel Xeon Gold 6346 up to 3.6 GHz1960 GB * 2 EA
h2v32m512_metal16 Core32 vCore512 GBIntel Xeon Gold 6346 with up to 3.6 GHz1960 GB * 2 EA
h2v72m256_metal36 Core72 vCore256 GBIntel Xeon Gold 6354 up to 3.6 GHz21.92 TB * 2 EA
h2v72m512_metal36 Core72 vCore512 GBIntel Xeon Gold 6354 up to 3.6 GHz21.92 TB * 2 EA
h2v72m1024_metal36 Core72 vCore1,024 GBIntel Xeon Gold 6354 up to 3.6 GHz21.92 TB * 2 EA
h2v96m384_metal48 Core96 vCore384 GBIntel Xeon Gold 6342 up to 3.3 GHz21.92 TB * 2 EA
h2v96m768_metal48 Core96 vCore768 GBIntel Xeon Gold 6342 up to 3.3 GHz21.92 TB * 2 EA
h2v96m1536_metal48 Core96 vCore1,536 GBIntel Xeon Gold 6342 with a maximum of 3.3 GHz21.92 TB * 2 EA
표. Bare Metal Server 서버 타입 사양 > s2/h2 서버 타입

Supported server types by OS Image

The server type varies depending on the OS image provided by the Bare Metal Server. The supported server types for each OS image are as follows.

OS Image versions4/h4 support statusWhether s3/h3 is supportedWhether s2/h2 is supported
Oracle Linux 9.6OOO
RHEL 8.10XOO
RHEL 9.4OOO
RHEL 9.6OOO
Rocky Linux 8.10XOO
Rocky Linux 9.6OOO
Ubuntu 22.04Verification neededOO
Ubuntu 24.04OOVerification required
Windows 2019XOO
Windows 2022OOO
표. OS Image별 지원 서버 타입
참고
  • s3/h3, s2/h2 server types have closed new service applications. * There is no impact on the services currently in use.
  • For Ubuntu, the server types you can request may vary depending on resource availability. * Please verify before applying for the service.

4.1.2 - Monitoring Metrics

Cloud Monitoring service termination notice

According to Samsung Cloud Platform’s policy, the Cloud Monitoring service is scheduled to be discontinued in September 2026.
Accordingly, after the September 2026 release, resource monitoring of the Samsung Cloud Platform via Cloud Monitoring will no longer be possible.

With the new alternative service, you can continuously perform resource monitoring by leveraging ServiceWatch released in October 2025.
ServiceWatch provides more modern and powerful features, replacing Cloud Monitoring to deliver a seamless monitoring environment.

If you are collecting metrics and logs through the Cloud Monitoring Agent, you need to switch to the ServiceWatch Agent.

For detailed information about ServiceWatch, please refer to ServiceWatch Overview.
Detailed information about ServiceWatch Agent can be found in the ServiceWatch Agent.

Bare Metal Server Monitoring Metrics

The table below shows the monitoring metrics of Bare Metal Server that can be viewed through Cloud Monitoring.

Guide
Bare Metal Server requires the user to install the Agent through the guide in order to view monitoring metrics. Before using the reliable Bare Metal Server service, please be sure to install the Agent. For instructions on installing the Agent and detailed usage of Cloud Monitoring, refer to the Cloud Monitoring guide.
Performance itemsDetailed descriptionunit
Core Usage [IO Wait]Ratio of CPU time spent in wait state (disk wait)%
Core Usage [System]Proportion of CPU time spent in kernel space%
Core Usage [User]Proportion of CPU time spent in user space%
CPU CoresNumber of CPU cores on the hostcnt
CPU Usage [Active]Percentage of CPU time used, excluding Idle and IOWait states%
CPU Usage [Idle]It is the proportion of CPU time spent in idle state.%
CPU Usage [IO Wait]Ratio of CPU time spent in wait state (disk wait)%
CPU Usage [System]Percentage of CPU time used by the kernel%
CPU Usage [User]Percentage of CPU time used in user space%
CPU Usage/Core [Active]Percentage of CPU time used other than Idle and IOWait states%
CPU Usage/Core [Idle]Ratio of CPU time spent in idle state%
CPU Usage/Core [IO Wait]Ratio of CPU time spent in wait state (disk wait)%
CPU Usage/Core [System]Percentage of CPU time used by the kernel%
CPU Usage/Core [User]Percentage of CPU time used in user space%
Disk CPU Usage [IO Request]Proportion of CPU time during which I/O requests to the device were executed%
Disk Queue Size [Avg]Average queue length of requests executed for the devicenum
Disk Read BytesBytes per second read from the devicebytes
Disk Read Bytes [Delta Avg]Average of system.diskio.read.bytes_delta for individual disksbytes
Disk Read Bytes [Delta Max]Maximum system.diskio.read.bytes_delta of individual disksbytes
Disk Read Bytes [Delta Min]minimum system.diskio.read.bytes_delta of individual disksbytes
Disk Read Bytes [Delta Sum]Sum of system.diskio.read.bytes_delta of individual disksbytes
Disk Read Bytes [Delta]Delta of the system.diskio.read.bytes value for each Diskbytes
Disk Read Bytes [Success]Total bytes successfully readbytes
Disk Read RequestsNumber of read requests to the disk device per secondcnt
Disk Read Requests [Delta Avg]Average of system.diskio.read.count_delta for individual diskscnt
Disk Read Requests [Delta Max]Maximum system.diskio.read.count_delta for individual diskscnt
Disk Read Requests [Delta Min]Minimum of individual Disks’ system.diskio.read.count_deltacnt
Disk Read Requests [Delta Sum]Sum of system.diskio.read.count_delta for individual diskscnt
Disk Read Requests [Success Delta]Delta of system.diskio.read.count for each Diskcnt
Disk Read Requests [Success]Total number of successful readscnt
Disk Request Size [Avg]Average size of requests executed on the device (unit: sectors)num
Disk Service Time [Avg]Average service time (ms) of input requests executed on the devicems
Disk Wait Time [Avg]Average time taken for requests executed on the device to be supportedms
Disk Wait Time [Read]Average disk wait timems
Disk Wait Time [Write]Disk average wait timems
Disk Write Bytes [Delta Avg]Average of system.diskio.write.bytes_delta for individual disksbytes
Disk Write Bytes [Delta Max]Maximum system.diskio.write.bytes_delta of individual disksbytes
Disk Write Bytes [Delta Min]Minimum of system.diskio.write.bytes_delta for individual disksbytes
Disk Write Bytes [Delta Sum]Sum of system.diskio.write.bytes_delta of individual disksbytes
Disk Write Bytes [Delta]individual Disk’s system.diskio.write.bytes value deltabytes
Disk Write Bytes [Success]Total bytes successfully writtenbytes
Disk Write RequestsNumber of write requests to the disk device per secondcnt
Disk Write Requests [Delta Avg]Average of system.diskio.write.count_delta for individual diskscnt
Disk Write Requests [Delta Max]Maximum system.diskio.write.count_delta of individual diskscnt
Disk Write Requests [Delta Min]Minimum of system.diskio.write.count_delta for individual diskscnt
Disk Write Requests [Delta Sum]Sum of system.diskio.write.count_delta of individual diskscnt
Disk Write Requests [Success Delta]Delta of system.diskio.write.count for each Diskcnt
Disk Write Requests [Success]Total number of successful writescnt
Disk Writes BytesBytes per second written to the devicebytes
Filesystem Hang Checkfilesystem(local/NFS) hang check (normal:1, abnormal:0)status
Filesystem NodesTotal number of file nodes in the file systemcnt
Filesystem Nodes [Free]Total number of available file nodes in the file systemcnt
Filesystem Size [Available]Disk space (bytes) available to unauthorized usersbytes
Filesystem Size [Free]Available disk space (bytes)bytes
Filesystem Size [Total]Total disk space (bytes)bytes
Filesystem UsageUsed disk space percentage%
Filesystem Usage [Avg]Average of individual filesystem.used.pct%
Filesystem Usage [Inode]inode usage%
Filesystem Usage [Max]Maximum among individual filesystem.used.pct%
Filesystem Usage [Min]minimum among individual filesystem.used.pct%
Filesystem Usage [Total]Total filesystem usage%
Filesystem UsedUsed disk space (bytes)bytes
Filesystem Used [Inode]inode usagebytes
Memory FreeTotal amount of available memory (bytes). Memory used by system cache and buffers is not included.bytes
Memory Free [Actual]Actual usable memory (bytes).bytes
Memory Free [Swap]Available swap memory.bytes
Memory Totaltotal memorybytes
Memory Total [Swap]Total swap memory.bytes
Memory UsagePercentage of used memory%
Memory Usage [Actual]Percentage of memory actually used%
Memory Usage [Cache Swap]cached swap usage%
Memory Usage [Swap]Percentage of used swap memory%
Memory Usedused memorybytes
Memory Used [Actual]Actual used memory (bytes). The value obtained by subtracting used memory from total memory.bytes
Memory Used [Swap]Used swap memorybytes
CollisionsNetwork collisioncnt
Network In BytesNumber of received bytesbytes
Network In Bytes [Delta Avg]Average of system.network.in.bytes_delta for each networkbytes
Network In Bytes [Delta Max]Maximum of system.network.in.bytes_delta for each networkbytes
Network In Bytes [Delta Min]Minimum of system.network.in.bytes_delta for each networkbytes
Network In Bytes [Delta Sum]Sum of system.network.in.bytes_delta for individual networksbytes
Network In Bytes [Delta]Delta of received byte countbytes
Network In DroppedNumber of deleted packets among incoming packetscnt
Network In ErrorsNumber of errors during receptioncnt
Network In PacketsNumber of received packetscnt
Network In Packets [Delta Avg]Average of system.network.in.packets_delta for each networkcnt
Network In Packets [Delta Max]Maximum of system.network.in.packets_delta for each networkcnt
Network In Packets [Delta Min]Minimum of system.network.in.packets_delta for each networkcnt
Network In Packets [Delta Sum]Sum of system.network.in.packets_delta for individual networkscnt
Network In Packets [Delta]Delta of received packet countcnt
Network Out BytesNumber of transmitted bytesbytes
Network Out Bytes [Delta Avg]Average of system.network.out.bytes_delta for each networkbytes
Network Out Bytes [Delta Max]Maximum of system.network.out.bytes_delta for individual networksbytes
Network Out Bytes [Delta Min]Minimum of system.network.out.bytes_delta for individual networksbytes
Network Out Bytes [Delta Sum]Sum of system.network.out.bytes_delta for individual networksbytes
Network Out Bytes [Delta]Delta of transmitted byte countbytes
Network Out DroppedNumber of deleted packets among outgoing packetscnt
Network Out ErrorsNumber of errors during transmissioncnt
Network Out PacketsNumber of transmitted packetscnt
Network Out Packets [Delta Avg]Average of system.network.out.packets_delta for each networkcnt
Network Out Packets [Delta Max]Maximum of system.network.out.packets_delta for each networkcnt
Network Out Packets [Delta Min]Minimum of system.network.out.packets_delta for each networkcnt
Network Out Packets [Delta Sum]Sum of system.network.out.packets_delta for individual networkscnt
Network Out Packets [Delta]Delta of transmitted packet countcnt
Open Connections [TCP]All open TCP connectionscnt
Open Connections [UDP]All open UDP connectionscnt
Port UsageAvailable port usage rate%
SYN Sent SocketsNumber of sockets in SYN_SENT state (when connecting from local to remote)cnt
Kernel PID Maxkernel.pid_max valuecnt
Kernel Thread Maxkernel.threads-max valuecnt
Process CPU UsagePercentage of CPU time consumed by the process since the last update%
Process CPU Usage/CorePercentage of CPU time used by the process since the last event%
Process Memory UsageThe proportion of main memory (RAM) occupied by a process%
Process Memory UsedResident Set size. The amount of memory a process occupies in RAM.bytes
Process PIDprocess pidpid
Process PPIDParent process PIDpid
Processes [Dead]Number of dead processescnt
Processes [Idle]Number of idle processescnt
Processes [Running]Number of running processescnt
Processes [Sleeping]sleeping processes countcnt
Processes [Stopped]number of stopped processescnt
Processes [Total]Total number of processescnt
Processes [Unknown]Number of processes with an unknown or unsearchable statuscnt
Processes [Zombie]zombie processes countcnt
Running Process Usageprocess usage%
Running ProcessesNumber of running processescnt
Running Thread UsageThread usage%
Running ThreadsTotal number of threads running in running processescnt
Context Switchescontext switch count (per second)cnt
Load/Core [1 min]The load over the last 1 minute divided by the number of corescnt
Load/Core [15 min]The load over the last 15 minutes divided by the number of corescnt
Load/Core [5 min]The load over the last 5 minutes divided by the number of corescnt
Multipaths [Active]External storage connection path status = active countcnt
Multipaths [Failed]External storage connection path status = failed countcnt
Multipaths [Faulty]External storage connection path status = faulty countcnt
NTP Offsetmeasured offset of the last sample (the time difference between the NTP server and the local environment)num
Run Queue LengthExecution queue lengthnum
UptimeOS uptime (uptime). (milliseconds)ms
Context SwitchiesCPU context switch count (per second)cnt
Disk Read Bytes [Sec]Bytes read per second from the Windows logical diskcnt
Disk Read Time [Avg]Average data read time (seconds)sec
Disk Transfer Time [Avg]Disk average wait timesec
Disk UsageDisk usage%
Disk Write Bytes [Sec]Number of bytes written in one second on a Windows logical diskcnt
Disk Write Time [Avg]Average data write time (seconds)sec
Pagingfile UsagePaging file usage%
Pool Used [Non Paged]Nonpaged Pool usage in kernel memorybytes
Pool Used [Paged]Paged Pool usage in kernel memorybytes
Process [Running]Number of currently running processescnt
Threads [Running]Current number of running threadscnt
Threads [Waiting]Number of threads waiting for processor timecnt
Table. Bare Metal Server monitoring metrics (available when the Agent is installed)

4.2 - How-to guides

Users can create the service by entering the required information for a Bare Metal Server and selecting detailed options through the Samsung Cloud Platform Console.

Create Bare Metal Server

You can create and use the Bare Metal Server service from the Samsung Cloud Platform Console.

To create a Bare Metal Server, follow these steps.

  1. All Services > Compute > Bare Metal Server Click the menu. 1. Go to the Service Home page of the Bare Metal Server.
  2. On the Service Home page, click the Create Bare Metal Server button. 2. Go to the Bare Metal Server Creation page.
  3. Bare Metal Server Creation page, enter the information required to create the service and select detailed options.
    • Select the required information in the Image and version selection area.
      Category
      required status
      Detailed description
      ImageRequiredSelect the type of Image provided
      • RHEL, Rocky Linux, Ubuntu, Windows
      • Refer to the caution information below for the server types available for each OS Image
      Image versionRequiredSelect version of the selected Image
      • Provide a version list of the provided server Image
      Table. Bare Metal Server Image and version input items
      Caution

      The server types that can be requested vary by OS image as follows.

      OS Image versions4/h4 server types3/h3 server types2/h2 server type
      Oracle Linux 9.6OOO
      RHEL 8.10XOO
      RHEL 9.4OOO
      RHEL 9.6OOO
      Rocky Linux 8.10XOO
      Rocky Linux 9.6OOO
      Ubuntu 22.04Confirmation neededOO
      Ubuntu 24.04OOConfirmation needed
      Windows 2019XOO
      Windows 2022OOO
      • s3/h3, s2/h2 server types are no longer accepting new service applications. * There is no impact on the services currently in use.
      • For Ubuntu, the server types you can request may vary depending on resource availability. * Please confirm before applying for the service.
    • Enter or select the required information in the Service Information Input area.
      Category
      required status
      Detailed description
      Number of serversRequiredNumber of Bare Metal Servers to create simultaneously
      • Only numeric input is allowed, and up to 20 servers of the same type can be created simultaneously
      Service Type > Server TypeRequiredBare Metal Server server type
      • Select the desired vCPU, Memory, and Disk specifications
      Service Type > Planned ComputeRequiredStatus of resources with Planned Compute configured
      • In Use: Number of resources with Planned Compute that are currently in use
      • Configured: Number of resources with Planned Compute configured
      • Coverage Preview: Amount applied per resource by Planned Compute
      Table. Bare Metal Server Service Information Input Items
    • In the Required Information Input area, enter or select the required information.
      Category
      required status
      Detailed description
      Administrator accountRequiredSet the administrator account and password to be used when connecting to the server
      • RHEL and Ubuntu OS are provided with a fixed root account
      • For Windows OS, enter 5 to 20 characters using lowercase letters and numbers
        • Administrator Not allowed
      Server nameRequiredWhen the selected number of servers is 1, enter a name to distinguish the Bare Metal Server
      • It must start with a lowercase English letter and be entered using lowercase letters, numbers, and special characters (-) within 3 to 15 characters
      • It must not end with a special character (-)
      • Set the hostname to the entered server name
      Server name (auto-generated number)RequiredIt is enabled when the number of selected servers is 2 or more, and you must enter a name for distinguishing the Bare Metal Server in the format name-### (e.g., test-001)
      • It must start with a lowercase English letter and be entered using lowercase letters, numbers, and special characters (-) within 3 to 15 characters
      • It must not end with a special character (-)
      • Set the hostname to the entered server name
      • Automatically generate by incrementing from the entered number by 1 (excluding numbers already in use)
      Network SettingsRequiredSet the network where the Bare Metal Server will be installed
      • Select a pre-created VPC
      • General Subnet: Select a pre-created general Subnet
        • IP can be auto-generated or user-provided, and if input is selected, the user enters the IP manually
      • NAT: Available only when there is one server and the VPC is connected to an Internet Gateway
        • When checked, a NAT IP can be selected
      • NAT IP: Select a NAT IP
        • If no NAT IP is available to select, click the Create New button to generate a Public IP
        • Click the Refresh button to view and select the generated Public IP
        • Creating a Public IP incurs charges according to the Public IP pricing policy
      • Local Subnet (optional): Choose to use a local Subnet
        • It is not a required element for creating the service
        • Select a pre-created local Subnet
        • IP can be auto-generated or user-provided, and if input is selected, the user enters the IP manually
      Table. Bare Metal Server required information entry items
Caution

Please use a firewall, etc., to control traffic access for the Bare Metal Server. Security Group is not provided.
The firewall of a Bare Metal Server can only be used to control traffic between the Bare Metal Server and Virtual Server. To use the Firewall of a Bare Metal Server, follow the steps below.

  1. Separate the VPC of Bare Metal Server: Separate them so that the Bare Metal Server and Virtual Server do not use the same VPC.
  2. Transit Gateway Creation: Create a Transit Gateway.
    • The integration between the VPC of a Virtual Server and the VPC of a Bare Metal Server uses a Transit Gateway.
    • When creating a Transit Gateway integration in the VPC of a Bare Metal Server, you must also create the Bare Metal Server’s firewall.
  3. Firewall Rule Registration: Register the rule in the Firewall of the Bare Metal Server.
  1. On the Bare Metal Server Creation page, in the Additional Information Input area, enter or select the required information.

    Category
    required status
    Detailed description
    Local disk partitionSelectionSet whether to use the local disk partition
    • Up to 10 partitions can be created, including the root partition
    • Can use up to 90% of the total capacity
    • After checking Use, you can configure partition information
    • Configure root partition information
      • Partition type: flat, lvm selectable
      • Partition name: enter partition name
        • Can be entered only when partition type is lvm
        • Enter up to 15 characters starting with a letter and containing letters, numbers, and special characters (-_)
      • Partition size: enter at least 50 GB
      • Filesystem type: select according to the used image
        • For RHEL, Rocky Linux: xfs, ext4
        • For Ubuntu: ext4, xfs, btrfs
        • For SLES: btrfs, xfs, ext4
      • Mount point: start with special character / and include letters, numbers, and special characters (-_) up to 15 characters
        • Cannot be entered when Filesystem type is swap
      • Available capacity: 90% of the default disk capacity provided when selecting a server
        • When setting partition size, the remaining capacity is automatically calculated and displayed
        • The total partition disk size cannot exceed the available capacity
    • Configure additional partition information
      • Partition type: flat, lvm selectable
      • Partition name: enter partition name
        • Partition type can be entered only when it is lvm
        • Enter up to 15 characters starting with a letter and containing letters, numbers, and special characters (-_) up to 15 characters
      • Partition size: enter at least 1 GB
      • Filesystem type: select according to the used image>
        • For RHEL, Rocky Linux: xfs, ext4, swap
        • For Ubuntu: ext4, xfs, btrfs, swap
        • For SLES: btrfs, xfs, ext4, swap
      • Mount point: start with special character / and include letters, numbers, and special characters (-_) up to 15 characters
        • Cannot be entered when Filesystem type is swap
      • Available capacity: 90% of the default disk capacity provided when selecting a server
        • When setting partition size, the remaining capacity is automatically calculated and displayed
        • The total partition disk size cannot exceed the available capacity
    Placement GroupSelectionServers belonging to the same Placement group are distributed across different racks
    • Provides distributed placement for up to two servers belonging to the same Placement group
      • For distributed placement of three or more servers, add additional Placement groups
    • Applicable only at initial creation and cannot be modified afterward
    • If you terminate the last server in a Placement group, that Placement group is automatically deleted
    LockSelectionUsing a lock prevents actions caused by mistakes, such as terminating, starting, or stopping the server.
    Init ScriptSelectionScript to run when the server starts
    • The Init Script must be selected differently depending on the Image type
      • For Windows: Select Batch Script
      • For Linux: Shell Script
    Table. Bare Metal Server Additional Information Input Items

  2. Summary Check the detailed information and estimated charges generated in the panel, and click the Create button.

    • Once creation is complete, check the created resources on the Bare Metal Server List page.

Check detailed information of Bare Metal Server

The Bare Metal Server service allows you to view and edit the complete resource list and detailed information. The Bare Metal Server Details page consists of Details, Tags, Activity History tabs.

To view detailed information about the Bare Metal Server, follow these steps.

  1. All Services > Compute > Bare Metal Server Click the menu. 1. Go to the Service Home page of the Bare Metal Server.
  2. On the Service Home page, click the Bare Metal Server menu. 2. Go to the Bare Metal Server List page.
  3. On the Bare Metal Server List page, click the resource to view detailed information. 3. Go to the Bare Metal Server Details page.
    • Bare Metal Server Details page displays status information and additional feature information, and consists of Details, Tags, Operation History tabs.
      CategoryDetailed description
      Bare Metal Server statusStatus of the Bare Metal Server created by the user
      • Creating: State while the server is being created
      • Running:: State when creation is complete and the server is usable
      • Editing:: State while the IP is being changed
      • Unknown: Error state
      • Starting: State while the server is starting
      • Stopping: State while the server is stopping
      • Stopped: State when the server has stopped
      • Terminating: State while terminating
      • Terminated: State when termination is complete
      Server operation control buttonButton to change server status
      • Start: Start a stopped server
      • Stop: Stop a running server
      Service cancellationCancel service button
      Table. Bare Metal Server status information and additional functions

Detailed Information

On the Bare Metal Server List page, you can view the detailed information of the selected resource and, if needed, modify the information.

CategoryDetailed description
serviceservice name
Resource TypeResource Type
SRNUnique resource ID in Samsung Cloud Platform
  • In Bare Metal Server, it means the Bare Metal Server SRN
Resource nameResource name
  • In Bare Metal Server, it refers to the server name
Resource IDUnique resource ID in the service
ConstructorUser who created the service
Creation date and timeService creation date and time
ModifierUser who edited the service information
Modification date and timeDate and time the service information was modified
Server nameserver name
Image/VersionServer OS image and version
Server typeDisplay vCPU and memory information
Planned ComputeResource status with Planned Compute configured
LockIndicates whether Lock is enabled/disabled
  • When Lock is enabled, it prevents server termination/start/stop actions, avoiding accidental operations
  • If you need to change the Lock property value, click the Edit button to set it
CPU settingsWhen configuring the CPU, CPU specification information and current status
  • GPU specification information: Click the Edit button to enable/disable usage and change specifications
  • GPU status: only displays Error (Error) or Editing (Editing) status
    • If in Error state, Edit is possible
Placement GroupPlacement Group name
NetworkNetwork information of the Bare Metal Server
  • VPC, standard Subnet, IP and status, Public NAT IP and status, Private NAT IP and status
  • If you need to change the IP, click the Edit button to configure
Local SubnetLocal Subnet information of the Bare Metal Server
  • Local Subnet name, Local Subnet IP, Vlan ID, Interface Name
  • If you need to add a local Subnet, click the Add button to configure
Block StorageBlock Storage information connected to the server
  • Volume name, disk type, capacity, status
  • Click the Add button to go to the Block Storage creation screen
Init ScriptView the Init Script content entered during server creation
Table. Bare Metal Server detailed information tab items

Tag

Bare Metal Server list page lets you view the tag information of the selected resource, and you can add, modify, or delete it.

CategoryDetailed description
Tag listTag list
  • You can view the Key and Value information of the tag
  • Up to 50 tags can be added per resource
  • When entering a tag, you can search and select from the list of previously created Keys and Values
Table. Bare Metal Server Tag tab items

Job History

Bare Metal Server List page allows you to view the operation history of the selected resource.

CategoryDetailed description
Task History ListResource change history
  • Operation timestamp, resource ID, resource name, operation details, event topic, operation result, check operator information
Table. Bare Metal Server Work History Tab Detailed Information Items

Bare Metal Server Resource Management

If you need server control and management functions for the created Bare Metal Server resources, you can perform tasks on the Bare Metal Server List or Bare Metal Server Details page.

Bare Metal Server Control Operation

You can start, stop, and restart a running Bare Metal Server.

Notice
  • When a Bare Metal Server is stopped, the server’s power is turned off.
  • Since stopping may affect applications or storage in use, it is recommended to shut down the OS before stopping the Bare Metal Server.
  • After shutting down the OS, be sure to also stop the console.
Caution
When starting a Bare Metal Server, be sure to start it from the Console.

To control the operation of a Bare Metal Server, follow these steps.

  1. All Services > Compute > Bare Metal Server Click the menu. 1. Go to the Service Home page of the Bare Metal Server.
  2. On the Service Home page, click the Bare Metal Server menu. 2. Go to the Bare Metal Server List page.
  3. Bare Metal Server List page, after selecting multiple servers, you can control multiple servers simultaneously using the Start and Stop buttons at the top of the table.
    • Bare Metal Server Details The page also allows you to start and stop the server.
  4. On the Bare Metal Server List page, click the resource to control its operation and navigate to the Bare Metal Server Detail page.
  5. Check the server status and click the server operation control button to complete the change.
    • Start: Start the stopped server.
    • Stop: Stops the running server.
Operation control not possible
  • If a Bare Metal Server cannot be started when a start request is made, refer to the following.
    • When Lock is enabled: Change the Lock setting to disabled, then try again.
    • If the Bare Metal Server’s status is not Stopped: Change the Bare Metal Server’s status to Stopped, then try again.
  • If a stop request for a Bare Metal Server cannot be performed, refer to the following.
    • If Lock is set: Change the Lock setting to disabled, then try again.
    • If the Bare Metal Server’s status is not Running: Change the Bare Metal Server’s status to Running, then try again.

Add Block Storage

You can add Block Storage to a Bare Metal Server.

To add Block Storage, follow the steps below.

  1. All Services > Compute > Bare Metal Server Click the menu. 1. Go to the Service Home page of the Bare Metal Server.
  2. On the Service Home page, click the Bare Metal Server menu. 2. Go to the Bare Metal Server List page.
  3. Bare Metal Server List page, click the server to which you want to add Block Storage. 3. Go to the Bare Metal Server Details page.
  4. On the Bare Metal Server Details page, click the Add button in the Block Storage section.
  5. When the popup confirming Block Storage addition opens, click the Confirm button. 5. Block Storage (BM) Creation Navigate to the page.
  6. On the Block Storage(BM) Creation page, enter the information required to create the service and create a Block Storage.
  7. Navigate to the Bare Metal Server Details page where Block Storage was added and confirm that Block Storage has been added.
Caution
After creating a Block Storage, you cannot increase its capacity.

Configure CPU

You can configure the CPU of a Bare Metal Server. To configure the CPU, follow these steps.

  1. All Services > Compute > Bare Metal Server Click the menu. 1. Go to the Service Home page of the Bare Metal Server.

  2. On the Service Home page, click the Bare Metal Server menu. 2. Go to the Bare Metal Server List page.

  3. On the Bare Metal Server List page, click the resource to configure the CPU. 3. Go to the Bare Metal Server Details page.

  4. On the Bare Metal Server Details page, click the Edit button in the CPU Settings item. 4. Edit CPU Settings A popup window opens.

  5. After entering whether to use the CPU settings and the configuration information, click the Confirm button.

    Category
    required status
    Detailed description
    Use statusRequiredSelect whether to enable the CPU setting
    • When Use is set, Thread per core and core count are required inputs
    Threads per coreRequiredSet Hyper threading usage
    • To disable Hyper threading, set Thread per core to 1
    Number of coresRequiredSet the number of vCPUs to activate
    • Changing the number of cores does not affect the price.
    Total vCPU count-The final number of vCPUs to be activated on the server
    • Automatically calculate and display the value obtained by multiplying Thread per core by number of cores
    Table. CPU configuration information items

  6. When the pop-up notifying CPU setting changes opens, click the Confirm button.

  7. Check that the CPU settings have been correctly modified on the Bare Metal Server Details page.

Caution
  • If you modify the CPU settings, they are reflected in the Console immediately, but they take effect on the actual server only after a reboot.
    • After modifying the settings, be sure to reboot (stop/start in the console) the server directly.
    • If you modify the CPU settings again before rebooting, an error occurs.
  • The number of configurable cores may vary depending on the server environment.
    • By default, it provides the option to select 25%, 50%, or 75% of the total cores.
    • Since the number of cores that can be configured when requesting a server may vary depending on the server environment, if you need any additional settings, be sure to inquire before applying for a Bare Metal Server and then proceed with the application.

Terminate Bare Metal Server

If you terminate an unused Bare Metal Server, you can reduce operating costs. However, if you terminate a Bare Metal Server, the running service may be stopped immediately, so you should proceed with the termination only after fully considering the impact of service interruption.

Caution
Please be aware that data cannot be recovered after terminating the service.
Caution
If you terminate servers one by one that have Block Storage (BM) attached, the servers will be terminated but the attached Block Storage (BM) will not be terminated, so terminate the Block Storage (BM) directly.

To cancel a Bare Metal Server, follow the steps below.

  1. All Services > Compute > Bare Metal Server Click the menu. 1. Go to the Service Home page of the Bare Metal Server.
  2. On the Service Home page, click the Bare Metal Server menu. 2. Go to the Bare Metal Server List page.
  3. On the Bare Metal Server List page, select the resource to terminate, and click the Terminate Service button.
    • You can select multiple resources and delete them simultaneously.
    • You can also delete the resource by clicking the Cancel Service button on the Bare metal Server Details page.
  4. Once the termination is complete, check the Bare Metal Server List page to see if the resource has been terminated.

Termination constraints

If a Bare Metal Server termination request cannot be completed, a popup will provide guidance. Please refer to the case below.

Cancellation not possible
  • When Block Storage(BM) is connected (simultaneous termination of two or more servers): Please disconnect the Block Storage(BM) first.

  • If File Storage is connected: First, disconnect the File Storage.

  • If Lock is set: Change the Lock setting to disabled, then try again.

  • If there are resources connected to Backup Agent or Load Balancer: First disconnect the connections of those resources.

  • If resource management tasks for Bare Metal Server are in progress on the same account: After the Bare Metal Server resource management tasks are completed, please try again.

  • If the Bare Metal Server’s status is not Running or Stopped: Change the Bare Metal Server’s status to Running or Stopped, then try again.

  • If the selection includes a server that cannot be terminated simultaneously: Select only resources that can be terminated, then try again.

Configure local Subnet

After completing the creation of a Bare Metal Server, when adding a local Subnet on the Bare Metal Server Details page, the user must manually configure the network settings of the local Subnet.

First connection (kr-west)

If there is no local subnet connected to the Bare Metal Server and you are adding the first connection, follow the guide below.

Caution

This guide applies to kr-west(Korea West) when adding the first local Subnet connection to the server.

Linux - Setting up Subnet on Ubuntu

To add a local Subnet and configure the network on an Ubuntu operating system, follow these steps.

  1. On the Bare Metal Server Details page, check the Interface Name.

  2. Retrieve the network configuration information.

    Color mode
    [root@localhost ~]# vi /etc/netplan/50-cloud-init.yaml
    network:
    ethernets:
     ens2f1:
     match:
     macaddress: 68:05:ca:d4:09:91
     mtu: 1500
     set-name: ens2f1
     ens4f1:
     match:
     macaddress: 68:05:ca:d4:09:01
     mtu: 1500
     set-name: ens4f1
    [root@localhost ~]# vi /etc/netplan/50-cloud-init.yaml
    network:
    ethernets:
     ens2f1:
     match:
     macaddress: 68:05:ca:d4:09:91
     mtu: 1500
     set-name: ens2f1
     ens4f1:
     match:
     macaddress: 68:05:ca:d4:09:01
     mtu: 1500
     set-name: ens4f1
    Code block. View network configuration file

  3. After adding a new VLAN, set the IP for the bonding configuration.

    • Replace the example code’s ID and IP with the assigned ID and IP.
      Color mode
      [root@localhost ~]# vi /etc/netplan/50-cloud-init.yaml
      network:
          bonds:
              bond-mgt:
                  interfaces:
                  - ens2f1      // **Bare Metal Server Details** Enter the Interface Name displayed on the page.
                  - ens4f1      // **Bare Metal Server Details** Enter the Interface Name shown on the page.
                  mtu: 1500
                  parameters:
                      mii-monitor-interval: 100
                      mode: active-backup
                      transmit-hash-policy: layer2
          ethernets:
              ens2f1:
                  match:
                  macaddress: 68:05:ca:d4:09:91
                  mtu: 1500
                  set-name: ens2f1
              ens4f1:
                  match:
                  macaddress: 68:05:ca:d4:09:01
                  mtu: 1500
                  set-name: ens4f1
          vlans:
              bond-mgt.20:   // Enter the Vlan ID you verified in the SCP Console instead of 20.
              addresses:
                  - 192.168.0.10/24 // Set it to the local Subnet IP confirmed in the SCP Console.
                  id: 20    // Set to the Vlan ID verified in the SCP Console.
                  link: bond-mgt
                  mtu: 1500
      [root@localhost ~]# vi /etc/netplan/50-cloud-init.yaml
      network:
          bonds:
              bond-mgt:
                  interfaces:
                  - ens2f1      // **Bare Metal Server Details** Enter the Interface Name displayed on the page.
                  - ens4f1      // **Bare Metal Server Details** Enter the Interface Name shown on the page.
                  mtu: 1500
                  parameters:
                      mii-monitor-interval: 100
                      mode: active-backup
                      transmit-hash-policy: layer2
          ethernets:
              ens2f1:
                  match:
                  macaddress: 68:05:ca:d4:09:91
                  mtu: 1500
                  set-name: ens2f1
              ens4f1:
                  match:
                  macaddress: 68:05:ca:d4:09:01
                  mtu: 1500
                  set-name: ens4f1
          vlans:
              bond-mgt.20:   // Enter the Vlan ID you verified in the SCP Console instead of 20.
              addresses:
                  - 192.168.0.10/24 // Set it to the local Subnet IP confirmed in the SCP Console.
                  id: 20    // Set to the Vlan ID verified in the SCP Console.
                  link: bond-mgt
                  mtu: 1500
      Code block. IP configuration
  4. Apply the changes to the system.

    Color mode
    # netplan apply
    # netplan apply
    Code block. Apply changes

  5. Check the interface status.

    Color mode
    # ip a
    or
    # bash /usr/local/bin/ip.sh
    # ip a
    or
    # bash /usr/local/bin/ip.sh
    Code block. Interface lookup

Linux – Configuring Subnet on CentOS/Red Hat

After adding a local Subnet on a CentOS/Red Hat operating system, follow these steps to configure the network.

Caution
Be careful, as setting the interface name incorrectly may delete the IP information currently in use.
  1. On the Bare Metal Server Details page, check the Interface Name.

  2. Modify the following command and execute it.

    Color mode
    #!/usr/bin/bash
    
    IP_ADDR="10.1.1.3/24"   // Set the local Subnet IP as observed in the Console.
    VLAN_ID="7"             // Set the Vlan ID as observed in the console.
    
    BOND_NAME="bond-mgt"
    BOND_IF_name1="ens2f1"  // Enter the Interface Name as found on the **Bare Metal Server Detailed** page.
    BOND_IF_name2="ens4f0"  // **Bare Metal Server Details** page, enter the Interface Name you verified.
    
    
    # Delete Connection
    nmcli con down "Bond ${BOND_NAME}"
    nmcli con del  "Bond ${BOND_NAME}"
    
    nmcli con down "System ${BOND_IF_name1}"
    nmcli con down "System ${BOND_IF_name2}"
    
    nmcli con del  "System ${BOND_IF_name1}"
    nmcli con del  "System ${BOND_IF_name2}"
    
    
    # Create Bonding
    nmcli con add con-name ${BOND_NAME} type bond ifname ${BOND_NAME}
    nmcli conn mod ${BOND_NAME} con-name "Bond ${BOND_NAME}"
    nmcli conn mod "Bond ${BOND_NAME}" ipv4.method    disabled
    nmcli conn mod "Bond ${BOND_NAME}" ipv6.method    ignore
    nmcli conn mod "Bond ${BOND_NAME}" connect.autoconnect yes
    
    nmcli conn mod "Bond ${BOND_NAME}" +bond.options mode=active-backup      \
                                       +bond.options xmit_hash_policy=layer2
                                       +bond.options miimon=100              \
                                       +bond.options num_grat_arp=1          \
                                       +bond.options downdelay=0
                                       +bond.options updelay=0
    
    # Assign bond-slave
    nmcli conn add type bond-slave ifname ${BOND_IF_name1}  con-name "${BOND_IF_name1}" master ${BOND_NAME}
    nmcli conn mod ${BOND_IF_name1} con-name "System ${BOND_IF_name1}"
    
    nmcli conn add type bond-slave ifname ${BOND_IF_name2}  con-name "${BOND_IF_name2}" master ${BOND_NAME}
    nmcli conn mod ${BOND_IF_name2} con-name "System ${BOND_IF_name2}"
    
    # Connection UP
    nmcli conn up   "Bond ${BOND_NAME}"
    # add vlan
    nmcli conn add type vlan ifname "${BOND_NAME}.${VLAN_ID}" con-name "${BOND_NAME}.${VLAN_ID}" id ${VLAN_ID} dev ${BOND_NAME}
    nmcli con  mod ${BOND_NAME}.${VLAN_ID} con-name "Vlan ${BOND_NAME}.${VLAN_ID}"
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" ipv4.addresses ${IP_ADDR}
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" ipv4.method manual
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" ipv6.method "ignore"
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" connect.autoconnect yes
    nmcli con  up  "Vlan ${BOND_NAME}.${VLAN_ID}"
    nmcli      device reapply ${BOND_NAME}.${VLAN_ID}
    #!/usr/bin/bash
    
    IP_ADDR="10.1.1.3/24"   // Set the local Subnet IP as observed in the Console.
    VLAN_ID="7"             // Set the Vlan ID as observed in the console.
    
    BOND_NAME="bond-mgt"
    BOND_IF_name1="ens2f1"  // Enter the Interface Name as found on the **Bare Metal Server Detailed** page.
    BOND_IF_name2="ens4f0"  // **Bare Metal Server Details** page, enter the Interface Name you verified.
    
    
    # Delete Connection
    nmcli con down "Bond ${BOND_NAME}"
    nmcli con del  "Bond ${BOND_NAME}"
    
    nmcli con down "System ${BOND_IF_name1}"
    nmcli con down "System ${BOND_IF_name2}"
    
    nmcli con del  "System ${BOND_IF_name1}"
    nmcli con del  "System ${BOND_IF_name2}"
    
    
    # Create Bonding
    nmcli con add con-name ${BOND_NAME} type bond ifname ${BOND_NAME}
    nmcli conn mod ${BOND_NAME} con-name "Bond ${BOND_NAME}"
    nmcli conn mod "Bond ${BOND_NAME}" ipv4.method    disabled
    nmcli conn mod "Bond ${BOND_NAME}" ipv6.method    ignore
    nmcli conn mod "Bond ${BOND_NAME}" connect.autoconnect yes
    
    nmcli conn mod "Bond ${BOND_NAME}" +bond.options mode=active-backup      \
                                       +bond.options xmit_hash_policy=layer2
                                       +bond.options miimon=100              \
                                       +bond.options num_grat_arp=1          \
                                       +bond.options downdelay=0
                                       +bond.options updelay=0
    
    # Assign bond-slave
    nmcli conn add type bond-slave ifname ${BOND_IF_name1}  con-name "${BOND_IF_name1}" master ${BOND_NAME}
    nmcli conn mod ${BOND_IF_name1} con-name "System ${BOND_IF_name1}"
    
    nmcli conn add type bond-slave ifname ${BOND_IF_name2}  con-name "${BOND_IF_name2}" master ${BOND_NAME}
    nmcli conn mod ${BOND_IF_name2} con-name "System ${BOND_IF_name2}"
    
    # Connection UP
    nmcli conn up   "Bond ${BOND_NAME}"
    # add vlan
    nmcli conn add type vlan ifname "${BOND_NAME}.${VLAN_ID}" con-name "${BOND_NAME}.${VLAN_ID}" id ${VLAN_ID} dev ${BOND_NAME}
    nmcli con  mod ${BOND_NAME}.${VLAN_ID} con-name "Vlan ${BOND_NAME}.${VLAN_ID}"
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" ipv4.addresses ${IP_ADDR}
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" ipv4.method manual
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" ipv6.method "ignore"
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" connect.autoconnect yes
    nmcli con  up  "Vlan ${BOND_NAME}.${VLAN_ID}"
    nmcli      device reapply ${BOND_NAME}.${VLAN_ID}
    Code block. IP configuration script

  3. Check the interface status.

    Color mode
    # ip a
    or
    # bash /usr/local/bin/ip.sh
    # ip a
    or
    # bash /usr/local/bin/ip.sh
    Code block. Interface lookup

Setting up Subnet in Windows

After adding a local Subnet on the Windows operating system, follow these steps to configure the network.

  1. Windows Start icon, right-click it, then run the Windows PowerShell(Administrator) program.

  2. On the Bare Metal Server Details page, check the Interface Name.

  3. Run ncpa.cpl from the Windows Run menu.

    Figure

  4. Check whether the interface is enabled, and enable it if necessary.

    • Activate the Interface Name identified on the Bare Metal Server Details page.
      Figure
  5. Create a Teaming.

    Color mode
    PS C:\> New-NetLbfoTeam Name bond-mgt TeamMembers ens2f1,ens4f1
    PS C:\> Set-NetLbfoTeam Name bond-mgt LoadBalancingAlgorithm Dynamic
    PS C:\> New-NetLbfoTeam Name bond-mgt TeamMembers ens2f1,ens4f1
    PS C:\> Set-NetLbfoTeam Name bond-mgt LoadBalancingAlgorithm Dynamic
    Code block. Create Teaming
    그림

  6. After adding a new VLAN, configure the IP.

    • Enter the VLAN ID and local Subnet IP confirmed on the Bare Metal Server Details page.
      Color mode
      PS C:\> Add-NetLbfoTeamNIC -Team bond_bond-mgt -VlanID 20 -Name bond-mgt.20
      PS C:\> Get-NetAdapter
      PS C:\> netsh interface ip set address bond-mgt.20 static 192.168.0.10/24
      PS C:\> Add-NetLbfoTeamNIC -Team bond_bond-mgt -VlanID 20 -Name bond-mgt.20
      PS C:\> Get-NetAdapter
      PS C:\> netsh interface ip set address bond-mgt.20 static 192.168.0.10/24
      Code block. Windows IP configuration
  7. In the Windows Run menu, execute ncpa.cpl to check the interface status.

First connection(kr-south)

If there is no local subnet connected to the Bare Metal Server and you are adding the first connection, follow the guide below.

Caution

This guide applies to kr-south(Korea region) when adding the first local Subnet connection to the server.

Linux - Setting up Subnet on Ubuntu

To add a local Subnet and configure the network on an Ubuntu operating system, follow these steps.

  1. After adding a new VLAN, set the IP.

    • Replace the example code’s ID and IP with the assigned ID and IP.
      Color mode
      [root@localhost ~]# vi /etc/netplan/50-cloud-init.yaml
      network:
          bonds:
              bond-mgt:
                  interfaces:
                  - ens2f1
                  - ens4f1
                  mtu: 1500
                  parameters:
                      mii-monitor-interval: 100
                      mode: active-backup
                      transmit-hash-policy: layer2
          ethernets:
              ens2f1:
                  match:
                  macaddress: 68:05:ca:d4:09:91
                  mtu: 1500
                  set-name: ens2f1
              ens4f1:
                  match:
                  macaddress: 68:05:ca:d4:09:01
                  mtu: 1500
                  set-name: ens4f1
          vlans:
              bond-mgt.20:
              addresses:
                  - 192.168.0.10/24
                  id: 20
                  link: bond-mgt
                  mtu: 1500
          vlans:
              bond-mgt.21: // Enter the VLAN ID you verified on the console instead of 21.
              addresses:
                  - 192.168.0.20/24 // Set it to the local Subnet IP verified in the console.
                  id: 21    // Set to the VLAN ID verified in the console.
                  link: bond-mgt
                  mtu: 1500
      [root@localhost ~]# vi /etc/netplan/50-cloud-init.yaml
      network:
          bonds:
              bond-mgt:
                  interfaces:
                  - ens2f1
                  - ens4f1
                  mtu: 1500
                  parameters:
                      mii-monitor-interval: 100
                      mode: active-backup
                      transmit-hash-policy: layer2
          ethernets:
              ens2f1:
                  match:
                  macaddress: 68:05:ca:d4:09:91
                  mtu: 1500
                  set-name: ens2f1
              ens4f1:
                  match:
                  macaddress: 68:05:ca:d4:09:01
                  mtu: 1500
                  set-name: ens4f1
          vlans:
              bond-mgt.20:
              addresses:
                  - 192.168.0.10/24
                  id: 20
                  link: bond-mgt
                  mtu: 1500
          vlans:
              bond-mgt.21: // Enter the VLAN ID you verified on the console instead of 21.
              addresses:
                  - 192.168.0.20/24 // Set it to the local Subnet IP verified in the console.
                  id: 21    // Set to the VLAN ID verified in the console.
                  link: bond-mgt
                  mtu: 1500
      code block. Add Vlan and set IP
  2. Apply the changes to the system.

    Color mode
    # netplan apply
    # netplan apply
    Code block. Reflect changes

  3. Check the interface status.

    Color mode
    # ip a
    or
    # bash /usr/local/bin/ip.sh
    # ip a
    or
    # bash /usr/local/bin/ip.sh
    Code block. Interface lookup

Linux – Configuring Subnet on CentOS/Red Hat

CentOS/Red Hat operating system, after adding a local Subnet, follow these steps to configure the network.

Caution
When adding a local Subnet or configuring the network incorrectly, be careful as the IP information in use may be deleted.
  1. Check the Bond Name for the local Subnet.
    Color mode
    # sh /usr/local/bin/ip.sh
    # sh /usr/local/bin/ip.sh
    code block. Verify bonding
  2. Modify the following command and execute it.
    Color mode
    #!/usr/bin/bash
    
    IP_ADDR="10.1.1.3/24"   // Set the local Subnet IP as observed in the Console.
    VLAN_ID="7"             // Set the Vlan ID as observed in the console.
    
    BOND_NAME="bond-mgt" // 1. Set the Bond Name you confirmed.
    
    # add vlan
    nmcli conn add type vlan ifname "${BOND_NAME}.${VLAN_ID}" con-name "${BOND_NAME}.${VLAN_ID}" id ${VLAN_ID} dev ${BOND_NAME}
    nmcli con  mod ${BOND_NAME}.${VLAN_ID} con-name "Vlan ${BOND_NAME}.${VLAN_ID}"
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" ipv4.addresses ${IP_ADDR}
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" ipv4.method manual
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" ipv6.method "ignore"
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" connect.autoconnect yes
    nmcli con  up  "Vlan ${BOND_NAME}.${VLAN_ID}"
    
    nmcli      device reapply ${BOND_NAME}.${VLAN_ID}
    #!/usr/bin/bash
    
    IP_ADDR="10.1.1.3/24"   // Set the local Subnet IP as observed in the Console.
    VLAN_ID="7"             // Set the Vlan ID as observed in the console.
    
    BOND_NAME="bond-mgt" // 1. Set the Bond Name you confirmed.
    
    # add vlan
    nmcli conn add type vlan ifname "${BOND_NAME}.${VLAN_ID}" con-name "${BOND_NAME}.${VLAN_ID}" id ${VLAN_ID} dev ${BOND_NAME}
    nmcli con  mod ${BOND_NAME}.${VLAN_ID} con-name "Vlan ${BOND_NAME}.${VLAN_ID}"
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" ipv4.addresses ${IP_ADDR}
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" ipv4.method manual
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" ipv6.method "ignore"
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" connect.autoconnect yes
    nmcli con  up  "Vlan ${BOND_NAME}.${VLAN_ID}"
    
    nmcli      device reapply ${BOND_NAME}.${VLAN_ID}
    Code block. IP configuration script
  3. Check the interface status.
    Color mode
    # ip a
    or
    # bash /usr/local/bin/ip.sh
    # ip a
    or
    # bash /usr/local/bin/ip.sh
    Code block. Interface lookup

Configuring Subnet on Windows

After adding a local Subnet on the Windows operating system, follow these steps to configure the network.

  1. Windows Start icon, right-click it, then run the Windows PowerShell(Administrator) program.

  2. Check the Teaming name for the local Subnet.

    Color mode
    PS C:\> Get-NetAdapter
    PS C:\> Get-NetAdapter
    code block. Check Windows interface

  3. After adding a new VLAN, set the IP.

    • Enter the Teaming name confirmed in step 2, and the Vlan ID and local Subnet IP confirmed in the Console.
      Color mode
      PS C:\> Add-NetLbfoTeamNIC -Team bond_bond-mgt -VlanID 20 -Name bond-mgt.20
      PS C:\> Get-NetAdapter
      PS C:\> netsh interface ip set address bond-mgt.20 static 192.168.0.10/24
      PS C:\> Add-NetLbfoTeamNIC -Team bond_bond-mgt -VlanID 20 -Name bond-mgt.20
      PS C:\> Get-NetAdapter
      PS C:\> netsh interface ip set address bond-mgt.20 static 192.168.0.10/24
      Code block. Create Teaming
  4. In the Windows Run menu, execute ncpa.cpl to check the interface status.

    Figure

Add second connection (kr-west, kr-south)

If there is a local subnet connected to the Bare Metal Server, the guide for the second additional connection is below.

Because a Bonding was already created when connecting the first local Subnet, there is no procedure to create Bonding when connecting the second local Subnet.

Please refer to the details below.

Notice
This guide can be applied commonly to kr-west and kr-south.

Linux - Setting up Subnet on Ubuntu

To add a local Subnet and configure the network on an Ubuntu operating system, follow these steps.

  1. After adding a new VLAN, set the IP.

    • Replace the example code’s ID and IP with the assigned ID and IP.
      Color mode
      [root@localhost ~]# vi /etc/netplan/50-cloud-init.yaml
      network:
          bonds:
              bond-mgt:
                  interfaces:
                  - ens2f1
                  - ens4f1
                  mtu: 1500
                  parameters:
                      mii-monitor-interval: 100
                      mode: active-backup
                      transmit-hash-policy: layer2
          ethernets:
              ens2f1:
                  match:
                  macaddress: 68:05:ca:d4:09:91
                  mtu: 1500
                  set-name: ens2f1
              ens4f1:
                  match:
                  macaddress: 68:05:ca:d4:09:01
                  mtu: 1500
                  set-name: ens4f1
          vlans:
              bond-mgt.20:
              addresses:
                  - 192.168.0.10/24
                  id: 20
                  link: bond-mgt
                  mtu: 1500
          vlans:
              bond-mgt.21: // Enter the VLAN ID you verified on the console instead of 21.
              addresses:
                  - 192.168.0.20/24 // Set it to the local Subnet IP verified in the console.
                  id: 21    // Set to the VLAN ID verified in the console.
                  link: bond-mgt
                  mtu: 1500
      [root@localhost ~]# vi /etc/netplan/50-cloud-init.yaml
      network:
          bonds:
              bond-mgt:
                  interfaces:
                  - ens2f1
                  - ens4f1
                  mtu: 1500
                  parameters:
                      mii-monitor-interval: 100
                      mode: active-backup
                      transmit-hash-policy: layer2
          ethernets:
              ens2f1:
                  match:
                  macaddress: 68:05:ca:d4:09:91
                  mtu: 1500
                  set-name: ens2f1
              ens4f1:
                  match:
                  macaddress: 68:05:ca:d4:09:01
                  mtu: 1500
                  set-name: ens4f1
          vlans:
              bond-mgt.20:
              addresses:
                  - 192.168.0.10/24
                  id: 20
                  link: bond-mgt
                  mtu: 1500
          vlans:
              bond-mgt.21: // Enter the VLAN ID you verified on the console instead of 21.
              addresses:
                  - 192.168.0.20/24 // Set it to the local Subnet IP verified in the console.
                  id: 21    // Set to the VLAN ID verified in the console.
                  link: bond-mgt
                  mtu: 1500
      code block. Add Vlan and set IP
  2. Apply the changes to the system.

    Color mode
    # netplan apply
    # netplan apply
    Code block. Reflect changes

  3. Check the interface status.

    Color mode
    # ip a
    or
    # bash /usr/local/bin/ip.sh
    # ip a
    or
    # bash /usr/local/bin/ip.sh
    Code block. Interface lookup

Linux – Configuring Subnet on CentOS/Red Hat

After adding a local Subnet on CentOS/Red Hat operating systems, follow the steps below to configure the network.

Caution
When adding a local Subnet or configuring the network incorrectly, be careful as the IP information in use may be deleted.
  1. Check the Bond Name for the local Subnet.
    Color mode
    # sh /usr/local/bin/ip.sh
    # sh /usr/local/bin/ip.sh
    code block. Verify bonding
  2. Modify the following command and execute it.
    Color mode
    #!/usr/bin/bash
    
    IP_ADDR="10.1.1.3/24"   // Set the local Subnet IP as observed in the Console.
    VLAN_ID="7"             // Set the Vlan ID as observed in the console.
    
    BOND_NAME="bond-mgt" // 1. Set the Bond Name you confirmed.
    
    # add vlan
    nmcli conn add type vlan ifname "${BOND_NAME}.${VLAN_ID}" con-name "${BOND_NAME}.${VLAN_ID}" id ${VLAN_ID} dev ${BOND_NAME}
    nmcli con  mod ${BOND_NAME}.${VLAN_ID} con-name "Vlan ${BOND_NAME}.${VLAN_ID}"
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" ipv4.addresses ${IP_ADDR}
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" ipv4.method manual
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" ipv6.method "ignore"
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" connect.autoconnect yes
    nmcli con  up  "Vlan ${BOND_NAME}.${VLAN_ID}"
    
    nmcli      device reapply ${BOND_NAME}.${VLAN_ID}
    #!/usr/bin/bash
    
    IP_ADDR="10.1.1.3/24"   // Set the local Subnet IP as observed in the Console.
    VLAN_ID="7"             // Set the Vlan ID as observed in the console.
    
    BOND_NAME="bond-mgt" // 1. Set the Bond Name you confirmed.
    
    # add vlan
    nmcli conn add type vlan ifname "${BOND_NAME}.${VLAN_ID}" con-name "${BOND_NAME}.${VLAN_ID}" id ${VLAN_ID} dev ${BOND_NAME}
    nmcli con  mod ${BOND_NAME}.${VLAN_ID} con-name "Vlan ${BOND_NAME}.${VLAN_ID}"
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" ipv4.addresses ${IP_ADDR}
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" ipv4.method manual
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" ipv6.method "ignore"
    nmcli con  mod "Vlan ${BOND_NAME}.${VLAN_ID}" connect.autoconnect yes
    nmcli con  up  "Vlan ${BOND_NAME}.${VLAN_ID}"
    
    nmcli      device reapply ${BOND_NAME}.${VLAN_ID}
    Code block. IP configuration script
  3. Check the interface status.
    Color mode
    # ip a
    or
    # bash /usr/local/bin/ip.sh
    # ip a
    or
    # bash /usr/local/bin/ip.sh
    Code block. Interface lookup

Configuring Subnet on Windows

After adding a local Subnet on the Windows operating system, follow these steps to configure the network.

  1. Windows Start icon, right-click it, then run the Windows PowerShell(Administrator) program.

  2. Check the Teaming name for the local Subnet.

    Color mode
    PS C:\> Get-NetAdapter
    PS C:\> Get-NetAdapter
    code block. Check Windows interface

  3. After adding a new VLAN, configure the IP.

    • Enter the Teaming name confirmed in step 2, the Vlan ID confirmed in the Console, and the local Subnet IP.
      Color mode
      PS C:\> Add-NetLbfoTeamNIC -Team bond_bond-mgt -VlanID 20 -Name bond-mgt.20
      PS C:\> Get-NetAdapter
      PS C:\> netsh interface ip set address bond-mgt.20 static 192.168.0.10/24
      PS C:\> Add-NetLbfoTeamNIC -Team bond_bond-mgt -VlanID 20 -Name bond-mgt.20
      PS C:\> Get-NetAdapter
      PS C:\> netsh interface ip set address bond-mgt.20 static 192.168.0.10/24
      Code block. Create Teaming
  4. In the Windows Run menu, execute ncpa.cpl to check the interface status.

    Figure

Change IP

The IP can be changed to carry out migration, server replacement, and similar tasks.

Caution
  • If you proceed with changing the IP, you will no longer be able to communicate using that IP, and you cannot cancel the IP change while it is in progress.
  • If the server is running the Load Balancer service, you must delete the old IP from the LB server group and directly add the new IP as a member of the LB server group.
  • Servers using Public NAT, Privat NAT must disable and reconfigure Public NAT, Privat NAT after an IP change.
    • If you are using Public NAT and Privat NAT, first disable the use of Public NAT and Privat NAT, complete the IP change, and then reconfigure.
    • Whether to use Public NAT and Privat NAT can be changed by clicking the Edit button for Public NAT IP and Privat NAT on the Bare Metal Server Details page.

Follow these steps to change the IP.

  1. All Services > Compute > Bare Metal Server Click the menu. 1. Go to the Service Home page of the Bare Metal Server.

  2. On the Service Home page, click the Bare Metal Server menu. 2. Go to the Bare Metal Server List page.

  3. Bare Metal Server List page, click the server whose IP you want to change. Navigate to the Bare Metal Server Detail page.

  4. On the Bare Metal Server Details page, click the Edit button next to the IP field.

  5. When the popup notifying IP modification opens, click the Confirm button. 5. IP Change The popup window opens.

  6. IP Change popup window’s Step 1, Step 2, Step 3 tasks should be performed sequentially.

    Notice
    • When changing the IP, the detailed configuration method for the IP change step varies depending on the subnet of the IP to be changed. * Be sure to refer to the following example and proceed with the tasks step by step.
    • When each step is completed successfully, the task status in the upper right corner is displayed as Completed, and you can proceed to the next step.
    • When performing the final check of Step 3, it is recommended to restart the server before proceeding with the inspection.

  7. After confirming that all tasks have completed successfully, click the Confirm button.

Change to an IP in the same Subnet

This explains how to configure IP settings per operating system when the IP to be changed uses the same subnet.

Linux – centos/redhat operating system

Step 1

Follow the steps below and proceed with Step 1.

  1. Select the Subnet to modify.
  2. Enter the IP to change.
  3. Click the IP Allocation Request button.
  4. When a pop-up notifying you of an IP change opens, click the Confirm button.
    • When the task completes successfully, the task status in the upper right corner will be displayed as Completed.
      Caution
      If you proceed with the IP allocation request of Step 1, you cannot cancel or restore the IP change.

Step 2

Follow the steps below to perform Step 2.

  1. For the IP change operation, connect to the target server using the NAT IP.

    Notice
    To prevent communication failures during operation, it is recommended to connect through another Virtual Server or Bare Metal Server created in the same subnet.

  2. Enter the assigned IP from Step 1 and set the new IP on the server.

    • In the following example, replace 172.17.34.150 with the assigned IP address.
    • After checking the information of the Interface you want to change on the server, replace it with bond-srv.9 in the following example.
      Color mode
      # nmcli con mod "Vlan bond-srv.9" ipv4.addresses 172.17.34.150/24
      # nmcli con mod "Vlan bond-srv.9" ipv4.method manual
      # nmcli  device reapply bond-srv.9
      # nmcli con mod "Vlan bond-srv.9" ipv4.addresses 172.17.34.150/24
      # nmcli con mod "Vlan bond-srv.9" ipv4.method manual
      # nmcli  device reapply bond-srv.9
      Code block. IP settings to modify
      Notice
      • Setting the IP disconnects the terminal session.
      • Step 2 After completing the task, if the task status changes to Completed, you can reconnect to the terminal.
  3. When all tasks are completed, select the task completion checkbox for Step 2 in the IP 변경 popup window.

    • When the task completes successfully, the task status in the upper right corner will be displayed as Completed.
      Notice
      • If the task status of Step 2 has changed to Completed but you still have issues with terminal access, go to the All Services > Management > Support Center Contact Us menu and contact us.

Step 3

Proceed with Step 3 according to the following procedure.

  1. Connect to the server to be changed using the NAT IP and check its communication status.

    • Use the following command to verify whether any pre‑change configuration information remains and whether the change was applied correctly. * If you can successfully connect to the server targeted for IP change, the changed IP is in normal communication state.
      Color mode
      # bash /usr/local/bin/ip.sh
      # bash /usr/local/bin/ip.sh
      Code block. Communication status check
      Reference
      The NAT IP does not change.
  2. After all tasks are completed, restart the server and then perform a final check.

    Reference
    It is recommended to perform the final check after restarting the server.

  3. If the final inspection shows no issues, select the work‑completion checkbox for Step 3 in the IP change popup.

Linux – Ubuntu operating system

Step 1

Follow the steps below and proceed with Step 1.

  1. Select the Subnet to modify.
  2. Enter the IP to change.
  3. Click the IP Allocation Request button.
  4. When a pop-up notifying you of an IP change opens, click the Confirm button.
    • When the task completes successfully, the task status in the upper right corner will be displayed as Completed.
      Caution
      If you proceed with the IP allocation request of Step 1, you cannot cancel or restore the IP change.

Step 2

Follow the steps below to perform Step 2.

  1. For the IP change operation, connect to the target server using the NAT IP.

    Notice
    To prevent communication failures during operation, it is recommended to connect through another Virtual Server or Bare Metal Server created in the same subnet.

  2. Enter the assigned IP from Step 1 and set the new IP on the server.

    • In the following example, replace 172.17.34.150/24 with the assigned IP address.
    • After checking the information of the Interface you want to change on the server, replace it with bond-srv.9 in the following example.
      Color mode
      [root@localhost ~]# vi /etc/netplan/50-cloud-init.yaml
      network:
          bonds:
          ......................     Omitted
          ethernets:
          ......................     Omitted
          vlans:
              bond-srv.9:
              addresses:
                  - 172.17.34.150/24   # Enter the IP assigned in Step1.
                  gateway4: 172.17.34.2
                  id: 9
                  link: bond-srv
                  mtu: 1500
              bond-srv.350:
                  addresses:
                  - 172.16.87.150/24
                  routes:
                  - to: 172.17.87.0/24
                    via: 172.16.87.1
      - to: 172.17.87.0/24
                    via: 172.16.87.1
                  id: 350
                  link: bond-srv
      [root@localhost ~]# vi /etc/netplan/50-cloud-init.yaml
      network:
          bonds:
          ......................     Omitted
          ethernets:
          ......................     Omitted
          vlans:
              bond-srv.9:
              addresses:
                  - 172.17.34.150/24   # Enter the IP assigned in Step1.
                  gateway4: 172.17.34.2
                  id: 9
                  link: bond-srv
                  mtu: 1500
              bond-srv.350:
                  addresses:
                  - 172.16.87.150/24
                  routes:
                  - to: 172.17.87.0/24
                    via: 172.16.87.1
      - to: 172.17.87.0/24
                    via: 172.16.87.1
                  id: 350
                  link: bond-srv
      Code block. IP settings to modify
  3. Use the Netplan apply command to apply the changes to the system.

    Color mode
    [root@localhost ~]# netplan apply
    [root@localhost ~]# netplan apply
    Code block. Run Netplan apply
    Notice
    • Setting the IP disconnects the terminal session.
    • Step 2 After completing the task, if the task status changes to Completed, you can reconnect to the terminal.

  4. When all tasks are completed, select the task completion checkbox for Step 2 in the IP Change popup.

    • When the task completes successfully, the task status in the upper right corner will be displayed as Completed.
      Notice
      • If the task status of Step 2 has changed to Completed but you still have issues with terminal access, go to the All Services > Management > Support Center Contact Us menu and submit an inquiry.

Step 3

Proceed with Step 3 according to the following procedure.

  1. Connect to the server to be changed using the NAT IP and check its communication status.

    • Use the following command to verify whether any pre‑change configuration information remains and whether the change was applied correctly. * If you can successfully connect to the server targeted for IP change, the new IP is operating normally.
      Color mode
      # bash /usr/local/bin/ip.sh
      # bash /usr/local/bin/ip.sh
      Code block. Communication status check
      Reference
      The NAT IP does not change.
  2. After all tasks are completed, restart the server and then perform a final check.

    Reference
    It is recommended to perform the final check after restarting the server.

  3. If there are no issues in the final inspection results, select the work completion checkbox for Step 3 in the IP Change popup window.

Windows operating system

Step 1

Follow the steps below and proceed with Step 1.

  1. Select the Subnet to modify.
  2. Enter the IP to change.
  3. Click the IP Allocation Request button.
  4. When the popup notifying IP change verification opens, click the Confirm button.
    • When the task completes successfully, the task status in the upper right corner will be displayed as Completed.
      Caution
      If you proceed with the IP allocation request of Step 1, you cannot cancel or restore the IP change.

Step 2

Follow the steps below to perform Step 2.

  1. For the IP change operation, connect to the target server using the NAT IP.

    Notice
    To prevent communication failures during operation, it is recommended to connect through another Virtual Server or Bare Metal Server created in the same subnet.

  2. Windows Start icon, right‑click it, then run Windows PowerShell(Administrator).

  3. Enter the assigned IP from Step 1 and set the new IP on the server.

    • In the following example, replace 172.17.34.150 with the assigned IP address.
      Color mode
      PS C:\> netsh interface ip set address "bond-srv.20" static 172.17.34.150 255.255.255.0
      PS C:\> netsh interface ip set address "bond-srv.20" static 172.17.34.150 255.255.255.0
      Code block. IP settings to modify
      Notice
      • Setting the IP disconnects the terminal session.
      • Step 2 After completing the task, if the task status changes to Completed, you can reconnect to the terminal.
  4. When all tasks are completed, select the task completion checkbox for Step 2 in the IP 변경 popup window.

    • When the task completes successfully, the task status in the upper right corner will be displayed as Completed.
      Notice
      • If the task status of Step 2 has changed to Completed but you still have issues with terminal access, go to the All Services > Management > Support Center Contact Us menu and contact us.

Step 3

Proceed with Step 3 according to the following procedure.

  1. Connect to the server to be changed using the NAT IP and check its communication status.

    • Use the following command to verify whether any pre‑change configuration information remains and whether the change was applied correctly. * If you can successfully connect to the server targeted for IP change, the changed IP is in normal communication state.
      Color mode
      PS C:\> Get-NetIPAddress | Format-Table
      PS C:\> Get-NetIPAddress | Format-Table
      Code block. Communication status check
      Reference
      The NAT IP does not change.
  2. After all tasks are completed, restart the server and then perform a final check.

    Reference
    It is recommended to perform the final check after restarting the server.

  3. If there are no issues in the final inspection results, select the work completion checkbox for Step 3 in the IP Change popup window.

Change to IP of another Subnet

This explains how to configure IP settings for each operating system when the IP to be changed uses a different subnet.

Linux – centos/redhat operating system

Step 1

Follow the steps below and proceed with Step 1.

  1. Select the Subnet to modify.
  2. Enter the IP to change.
  3. Click the IP Allocation Request button.
  4. When the popup notifying IP change verification opens, click the Confirm button.
    • When the operation completes successfully, Vlan ID check, Default Gateway check information is displayed, and the operation status in the upper right corner shows Completed.
      Caution
      If you proceed with the IP allocation request of Step 1, you cannot cancel or restore the IP change.

Step 2

Follow the steps below to perform Step 2.

  1. For the IP change operation, connect to the target server using the NAT IP.

    Notice
    To prevent communication failures during operation, it is recommended to connect through another Virtual Server or Bare Metal Server created in the same subnet.

  2. Add a new VLAN and configure the IP to add the IP to the server.

    • Add VLAN: Create the interface of the Vlan ID identified in Step 1. * In the following example, enter the assigned ID instead of 20.
    • IP configuration: Enter the IP assigned in Step 1. * In the following example, replace 192.168.0.10/24 with the assigned IP address.
      Color mode
      # nmcli conn add type vlan ifname "bond-srv.20" con-name "bond-srv.20" id 20 dev bond-srv
      # nmcli con mod bond-srv.20 con-name "Vlan bond-srv.20"
      # nmcli con mod "Vlan bond-srv.20" ipv4.addresses 192.168.0.10/24
      # nmcli con mod "Vlan bond-srv.20" ipv4.method manual
      # nmcli con mod "Vlan bond-srv.20" ipv6.method "ignore"
      # nmcli con up  "Vlan bond-srv.20"
      # nmcli conn add type vlan ifname "bond-srv.20" con-name "bond-srv.20" id 20 dev bond-srv
      # nmcli con mod bond-srv.20 con-name "Vlan bond-srv.20"
      # nmcli con mod "Vlan bond-srv.20" ipv4.addresses 192.168.0.10/24
      # nmcli con mod "Vlan bond-srv.20" ipv4.method manual
      # nmcli con mod "Vlan bond-srv.20" ipv6.method "ignore"
      # nmcli con up  "Vlan bond-srv.20"
      Code block. IP settings to modify
  3. Set the default gateway for the new VLAN.

    • Default gateway configuration: Enter the Default gateway IP assigned in Step 1. * In the following example, replace 192.168.0.1 with the assigned Default gateway IP.
      Color mode
      # nmcli con mod "Vlan bond-srv.20"  ipv4.gateway 192.168.0.1
      # nmcli device reapply bond-srv.20
      # nmcli con mod "Vlan bond-srv.20"  ipv4.gateway 192.168.0.1
      # nmcli device reapply bond-srv.20
      Code block. IP settings to modify
      Notice
      • If you set a Default Gateway on a new VLAN, the terminal session gets disconnected.
      • Step 2 After completing the task, if the task status changes to Completed, you can reconnect to the terminal.
  4. When all tasks are completed, select the task completion checkbox for Step 2 in the IP 변경 popup window.

    • When the task completes successfully, the task status in the upper right corner will be displayed as Completed.
      Notice
      • If the task status of Step 2 has changed to Completed but you still experience problems with terminal access, navigate to the All Services > Management > Support Center Contact Us menu and submit an inquiry.

Step 3

Proceed with Step 3 according to the following procedure.

  1. Connect to the server subject to IP change using a NAT IP.

  2. Verify the Default Gateway IP of the existing (pre-change) interface, then delete it.

    • In the following example, replace 192.168.10.1 with the IP you verified.
      Color mode
      # ip route del   default  via 192.168.10.1
      # ip route del   default  via 192.168.10.1
      Code block. Delete the Default Gateway IP of the existing interface
  3. Connect to the server to be changed using the NAT IP and check its communication status.

    • Use the following command to verify whether any pre‑change configuration information remains and whether the changes were applied correctly. * If you can successfully connect to the server targeted for IP change, the changed IP is in normal communication state.
      Color mode
      # netstat nr
      # bash /usr/local/bin/ip.sh
      # netstat nr
      # bash /usr/local/bin/ip.sh
      Code block. Communication status check
      Reference
      The NAT IP does not change.
  4. Check the VLAN information of the existing IP and then delete it from the server.

    • Enter the verified ID in place of 30 in the following example.
      Color mode
      # nmcli con delete "Vlan bond-srv.30"
      # nmcli con delete "Vlan bond-srv.30"
      Code block. Delete the VLAN information of the existing IP.
  5. After all tasks are completed, restart the server and then perform a final check.

    Reference
    It is recommended to perform the final check after restarting the server.

  6. If the final inspection results show no issues, select the work‑completion checkbox for Step 3 in the IP Change popup window.

Linux – Ubuntu operating system

Step 1

Follow the steps below and proceed with Step 1.

  1. Select the Subnet to modify.
  2. Enter the IP to change.
  3. Click the IP Allocation Request button.
  4. When a popup that notifies IP change verification opens, click the Confirm button.
    • When the operation completes successfully, Vlan ID check, Default Gateway check information is displayed, and the operation status in the upper right corner shows Completed.
      Caution
      If you proceed with the IP allocation request of Step 1, you cannot cancel or restore the IP change.

Step 2

Follow the steps below to perform Step 2.

  1. For the IP change operation, connect to the target server using the NAT IP.

    Notice
    To prevent communication failures during operation, it is recommended to connect through another Virtual Server or Bare Metal Server created in the same subnet.

  2. Add a new VLAN and set the IP and Default Gateway to add the IP to the server.

    • In the following example, this is the part where content is added at the bottom of the Step 1 task description.
    • In the following example, enter the assigned ID and IP instead of ID and IP.
      Color mode
      [root@localhost ~]# vi /etc/netplan/50-cloud-init.yaml
      network:
          bonds:
          ......................     Omitted
          ethernets:
          ......................     Omitted
          vlans:
              bond-srv.9:
              addresses:
                  - 172.17.34.150/24
      
                  gateway4: 172.17.34.2
                  id: 9
                  link: bond-srv
                  mtu: 1500
              bond-srv.350:
                  addresses:
                  - 172.16.87.150/24
                  routes:
                  - to: 172.17.87.0/24
                    via: 172.16.87.1
      - to: 172.17.87.0/24
                    via: 172.16.87.1
                  id: 350
                  link: bond-srv
      
      # Create the interface for the Vlan ID identified in Step1.
      # Enter the IP assigned in Step1.
      # Enter the Default gateway IP assigned in Step1.
      bond-srv.20:
              addresses:
                  - 192.168.0.10/24
                  gateway4: 192.168.0.1
                  id: 20
                  link: bond-srv
                  mtu: 1500
      [root@localhost ~]# vi /etc/netplan/50-cloud-init.yaml
      network:
          bonds:
          ......................     Omitted
          ethernets:
          ......................     Omitted
          vlans:
              bond-srv.9:
              addresses:
                  - 172.17.34.150/24
      
                  gateway4: 172.17.34.2
                  id: 9
                  link: bond-srv
                  mtu: 1500
              bond-srv.350:
                  addresses:
                  - 172.16.87.150/24
                  routes:
                  - to: 172.17.87.0/24
                    via: 172.16.87.1
      - to: 172.17.87.0/24
                    via: 172.16.87.1
                  id: 350
                  link: bond-srv
      
      # Create the interface for the Vlan ID identified in Step1.
      # Enter the IP assigned in Step1.
      # Enter the Default gateway IP assigned in Step1.
      bond-srv.20:
              addresses:
                  - 192.168.0.10/24
                  gateway4: 192.168.0.1
                  id: 20
                  link: bond-srv
                  mtu: 1500
      Code block. IP settings to modify
  3. Use the Netplan apply command to apply the changes to the system.

    Color mode
    [root@localhost ~]# netplan apply
    [root@localhost ~]# netplan apply
    Code block. Run Netplan apply
    Notice
    • If you set a new Default Gateway, the terminal session will be disconnected.
    • Step 2 After completing the task, if the task status changes to Completed, you can reconnect to the terminal.

  4. When all tasks are completed, select the task completion checkbox for Step 2 in the IP 변경 popup window.

    • When the task completes successfully, the task status in the upper right corner will be displayed as Completed.
      Notice
      • If the task status of Step 2 has changed to Completed but you still experience problems with terminal access, navigate to the All Services > Management > Support Center Contact Us menu and submit an inquiry.

Step 3

Proceed with Step 3 according to the following procedure.

  1. Connect to the server subject to IP change using a NAT IP.

  2. Verify the Default Gateway IP of the existing (pre-change) interface, then delete it.

    • In the following example, the Delete this line row is the part that gets deleted.
      Color mode
      [root@localhost ~]# vi /etc/netplan/50-cloud-init.yaml
      network:
          bonds:
          ......................     Omitted
          ethernets:
          ......................     Omitted
          vlans:
              bond-srv.9:
              addresses:
                  - 172.17.34.150/24
                  gateway4: 172.17.34.2    #Delete this line
                  id: 9
                  link: bond-srv
                  mtu: 1500
              bond-srv.350:
                  addresses:
                  - 172.16.87.150/24
                  routes:
                  - to: 172.17.87.0/24
                    via: 172.16.87.1
      - to: 172.17.87.0/24
                    via: 172.16.87.1
                  id: 350
                  link: bond-srv
      
      bond-srv.20:
              addresses:
                  - 192.168.0.10/24
                  gateway4: 192.168.0.1
                  id: 20
                  link: bond-srv
                  mtu: 1500
      [root@localhost ~]# vi /etc/netplan/50-cloud-init.yaml
      network:
          bonds:
          ......................     Omitted
          ethernets:
          ......................     Omitted
          vlans:
              bond-srv.9:
              addresses:
                  - 172.17.34.150/24
                  gateway4: 172.17.34.2    #Delete this line
                  id: 9
                  link: bond-srv
                  mtu: 1500
              bond-srv.350:
                  addresses:
                  - 172.16.87.150/24
                  routes:
                  - to: 172.17.87.0/24
                    via: 172.16.87.1
      - to: 172.17.87.0/24
                    via: 172.16.87.1
                  id: 350
                  link: bond-srv
      
      bond-srv.20:
              addresses:
                  - 192.168.0.10/24
                  gateway4: 192.168.0.1
                  id: 20
                  link: bond-srv
                  mtu: 1500
      Code block. Delete the Default Gateway IP of the existing interface
  3. Connect to the server to be changed using the NAT IP and check its communication status.

    • Use the following command to verify whether any pre‑change configuration information remains and whether the change was applied correctly. * If you can successfully connect to the server targeted for IP change, the changed IP is in normal communication state.
      Color mode
      # netstat nr
      # bash /usr/local/bin/ip.sh
      # netstat nr
      # bash /usr/local/bin/ip.sh
      Code block. Communication status check
      Reference
      The NAT IP does not change.
  4. Delete the existing IP.

    • In the following example, the Delete this line row is the part that gets deleted.
      Color mode
      [root@localhost ~]# vi /etc/netplan/50-cloud-init.yaml
      network:
          bonds:
          ......................     Omitted
          ethernets:
          ......................     Omitted
          vlans:
              bond-srv.9:                   #Delete this line
              addresses:                    #Delete this line
                  - 172.17.34.150/24        # Delete this line
                  gateway4: 172.17.34.2     #Delete this line
                  id: 9                     # Delete this line
                  link: bond-srv            #Delete this line
                  mtu: 1500                 #Delete this line
              bond-srv.350:
                  addresses:
                  - 172.16.87.150/24
                  routes:
                  - to: 172.17.87.0/24
                    via: 172.16.87.1
      - to: 172.17.87.0/24
                    via: 172.16.87.1
                  id: 350
                  link: bond-srv
      
      bond-srv.20:
              addresses:
                  - 192.168.0.10/24
                  gateway4: 192.168.0.1
                  id: 20
                  link: bond-srv
                  mtu: 1
      [root@localhost ~]# vi /etc/netplan/50-cloud-init.yaml
      network:
          bonds:
          ......................     Omitted
          ethernets:
          ......................     Omitted
          vlans:
              bond-srv.9:                   #Delete this line
              addresses:                    #Delete this line
                  - 172.17.34.150/24        # Delete this line
                  gateway4: 172.17.34.2     #Delete this line
                  id: 9                     # Delete this line
                  link: bond-srv            #Delete this line
                  mtu: 1500                 #Delete this line
              bond-srv.350:
                  addresses:
                  - 172.16.87.150/24
                  routes:
                  - to: 172.17.87.0/24
                    via: 172.16.87.1
      - to: 172.17.87.0/24
                    via: 172.16.87.1
                  id: 350
                  link: bond-srv
      
      bond-srv.20:
              addresses:
                  - 192.168.0.10/24
                  gateway4: 192.168.0.1
                  id: 20
                  link: bond-srv
                  mtu: 1
      Code block. Delete existing IP
  5. Apply the changes to the system.

    Color mode
    [root@localhost ~]# netplan apply
    [root@localhost ~]#  ip link delete bond-srv.9 # Additional steps when deleting VLAN
    [root@localhost ~]# netplan apply
    [root@localhost ~]#  ip link delete bond-srv.9 # Additional steps when deleting VLAN
    Code block. Apply changes

  6. After all tasks are completed, restart the server and then perform a final check.

Reference
It is recommended to perform the final check after restarting the server.
  1. If the final inspection results show no issues, select the work‑completion checkbox for Step 3 in the IP Change popup window.

Windows operating system

Step 1

Follow the steps below and proceed with Step 1.

  1. Select the Subnet to modify.
  2. Enter the IP to change.
  3. Click the IP Allocation Request button.
  4. When the popup indicating IP change verification opens, click the Confirm button.
    • When the operation completes successfully, Vlan ID check, Default Gateway check information is displayed, and the operation status in the upper right corner shows Completed.
      Caution
      If you proceed with the IP allocation request of Step 1, you cannot cancel or restore the IP change.

Step 2

Follow the steps below to perform Step 2.

  1. For the IP change operation, connect to the target server using the NAT IP.

    Notice
    To prevent communication failures during operation, it is recommended to connect through another Virtual Server or Bare Metal Server created in the same subnet.

  2. Windows Start icon, right-click, then run Windows PowerShell(Administrator).

  3. Add a VLAN and configure the IP and default gateway.

    • Add VLAN: Create the interface of the Vlan ID identified in Step 1. * In the following example, enter the assigned ID instead of 20.
    • IP configuration: Enter the IP assigned in Step 1. * In the following example, replace 46 with the ifindex obtained using Get‑NetAdapter, and replace 192.168.0.10 with the assigned IP address.
    • Default gateway configuration: Enter the Default gateway IP assigned in Step 1. * In the following example, replace 192.168.0.1 with the assigned Default gateway IP.
      Color mode
      PS C:\> Add-NetLbfoTeamNIC -Team bond_bond-srv -VlanID 20 -Name bond-srv.20 -Confirm:$false
      PS C:\> Get-NetAdapter
      PS C:\> New-NetIPAddress -InterfaceIndex 46 -IPAddress 192.168.0.10 -PrefixLength 24 defaultgateway 192.168.0.1
      PS C:\> Add-NetLbfoTeamNIC -Team bond_bond-srv -VlanID 20 -Name bond-srv.20 -Confirm:$false
      PS C:\> Get-NetAdapter
      PS C:\> New-NetIPAddress -InterfaceIndex 46 -IPAddress 192.168.0.10 -PrefixLength 24 defaultgateway 192.168.0.1
      Code block. IP settings to modify
      Notice
      • If you set a new Default Gateway, the terminal session will be disconnected.
      • Step 2 After completing the task, if the task status changes to Completed, you can reconnect to the terminal.
  4. When all tasks are completed, select the task completion checkbox for Step 2 in the IP 변경 popup window.

    • When the task completes successfully, the task status in the upper right corner will be displayed as Completed.
      Notice
      • If the task status of Step 2 has changed to Completed but you still experience problems with terminal access, navigate to the All Services > Management > Support Center Contact Us menu and submit an inquiry.

Step 3

Proceed with Step 3 according to the following procedure.

  1. Connect to the server subject to IP change using a NAT IP.

  2. Run the interface index (ifindex) to check the existing Default Gateway IP.

    Color mode
    PS C:\> Get-NetAdapter
    Name                      InterfaceDescription                    ifIndex Status       MacAddress             LinkSpeed
    ----                      --------------------                    ------- ------       ----------             ---------
    bond-srv.9                Microsoft Network Adapter Multiple...#4      30 Up           40-A6-B7-27-96-D5        50 Gbps
    bond-srv                  Microsoft Network Adapter Multiple...#3      19 Up           40-A6-B7-27-96-D5        50 Gbps
    bond-iscsi                Microsoft Network Adapter Multiple...#2      18 Up           40-A6-B7-27-96-D4        50 Gbps
    bond-backup               Microsoft Network Adapter Multiplexo... 22 Up           68-05-CA-C9-EB-88        20 Gbps
    eno2                      Intel(R) Ethernet Connection X722 fo... 12 Disabled     38-68-DD-36-A0-59         1 Gbps
    ens3f0                    Intel(R) Ethernet Network Adapter XX... 11 Up           40-A6-B7-27-96-D4        25 Gbps
    ………………………………………….. Omitted
    PS C:\> Get-NetAdapter
    Name                      InterfaceDescription                    ifIndex Status       MacAddress             LinkSpeed
    ----                      --------------------                    ------- ------       ----------             ---------
    bond-srv.9                Microsoft Network Adapter Multiple...#4      30 Up           40-A6-B7-27-96-D5        50 Gbps
    bond-srv                  Microsoft Network Adapter Multiple...#3      19 Up           40-A6-B7-27-96-D5        50 Gbps
    bond-iscsi                Microsoft Network Adapter Multiple...#2      18 Up           40-A6-B7-27-96-D4        50 Gbps
    bond-backup               Microsoft Network Adapter Multiplexo... 22 Up           68-05-CA-C9-EB-88        20 Gbps
    eno2                      Intel(R) Ethernet Connection X722 fo... 12 Disabled     38-68-DD-36-A0-59         1 Gbps
    ens3f0                    Intel(R) Ethernet Network Adapter XX... 11 Up           40-A6-B7-27-96-D4        25 Gbps
    ………………………………………….. Omitted
    code block. Run Get-NetAdapter
    Color mode
    PS C:\> get-netroute -ifindex 30
    
    ifIndex DestinationPrefix                              NextHop                                  RouteMetric PolicyStore
    ------- -----------------                              -------                                  ----------- -----------
    30      255.255.255.255/32                             0.0.0.0                                          256 ActiveStore
    30      224.0.0.0/4                                    0.0.0.0                                          256 ActiveStore
    30      172.17.35.0/24                                 172.17.35.1                                      256 ActiveStore
    30      172.17.34.255/32                               0.0.0.0                                          256 ActiveStore
    30      172.17.34.14/32                                0.0.0.0                                          256 ActiveStore
    30      172.17.34.0/24                                 0.0.0.0                                          256 ActiveStore
    30      0.0.0.0/0                                      172.17.34.1                                        1 ActiveStore
    PS C:\> get-netroute -ifindex 30
    
    ifIndex DestinationPrefix                              NextHop                                  RouteMetric PolicyStore
    ------- -----------------                              -------                                  ----------- -----------
    30      255.255.255.255/32                             0.0.0.0                                          256 ActiveStore
    30      224.0.0.0/4                                    0.0.0.0                                          256 ActiveStore
    30      172.17.35.0/24                                 172.17.35.1                                      256 ActiveStore
    30      172.17.34.255/32                               0.0.0.0                                          256 ActiveStore
    30      172.17.34.14/32                                0.0.0.0                                          256 ActiveStore
    30      172.17.34.0/24                                 0.0.0.0                                          256 ActiveStore
    30      0.0.0.0/0                                      172.17.34.1                                        1 ActiveStore
    Code block. Execute -ifindex

  3. Delete the existing Default Gateway IP.

    • In the following example, replace 30 with the ifindex obtained via Get-NetAdapter, and replace 172.17.34.1 with the IP you obtained.
      Color mode
      PS C:\> Remove-NetRoute -ifIndex 30 -DestinationPrefix 0.0.0.0/0 -NextHop 172.17.34.1 -Confirm:$false
      PS C:\> Remove-NetRoute -ifIndex 30 -DestinationPrefix 0.0.0.0/0 -NextHop 172.17.34.1 -Confirm:$false
      Code block. Delete Default Gateway IP
      Color mode
      PS C:\> get-netroute -ifindex 30
      
      ifIndex DestinationPrefix                              NextHop                                  RouteMetric PolicyStore
      ------- -----------------                              -------                                  ----------- -----------
      30      255.255.255.255/32                             0.0.0.0                                          256 ActiveStore
      30      224.0.0.0/4                                    0.0.0.0                                          256 ActiveStore
      30      172.17.34.255/32                               0.0.0.0                                          256 ActiveStore
      30      172.17.34.14/32                                0.0.0.0                                          256 ActiveStore
      30      172.17.34.0/24                                 0.0.0.0                                          256 ActiveStore
      PS C:\> get-netroute -ifindex 30
      
      ifIndex DestinationPrefix                              NextHop                                  RouteMetric PolicyStore
      ------- -----------------                              -------                                  ----------- -----------
      30      255.255.255.255/32                             0.0.0.0                                          256 ActiveStore
      30      224.0.0.0/4                                    0.0.0.0                                          256 ActiveStore
      30      172.17.34.255/32                               0.0.0.0                                          256 ActiveStore
      30      172.17.34.14/32                                0.0.0.0                                          256 ActiveStore
      30      172.17.34.0/24                                 0.0.0.0                                          256 ActiveStore
      Code block. Execute -ifindex
      Notice
      • If you delete the existing Default Gateway, the terminal session will be disconnected.
      • Step 2 After completing the task, if the task status changes to Completed, you can reconnect to the terminal.
  4. Connect to the server to be changed using the NAT IP and check its communication status.

    • Use the following command to verify whether any pre‑change configuration information remains and whether the change was applied correctly. * If you can successfully connect to the server targeted for IP change, the changed IP is in normal communication state.
      Color mode
      PS C:\> netstat nr | findstr Default
      PS C:\> Get-NetIPAddress | Format-Table
      PS C:\> netstat nr | findstr Default
      PS C:\> Get-NetIPAddress | Format-Table
      Code block. Communication status check
      Reference
      The NAT IP does not change.
  5. Check the VLAN information of the existing IP in the Team information.

    Color mode
    PS C:\> Get-NetLbfoTeam
    
    Name                   : bond_bond-srv
    Members                : {ens6f1, ens3f1}
    TeamNics               : {bond-srv, bond-srv.9}
    Team ingMode            : SwitchIndependent
    LoadBalancingAlgorithm : Dynamic
    Status                 : Up
    
    Name                   : bond_bond-iscsi
    Members                : {ens6f0, ens3f0}
    TeamNics               : bond-iscsi
    Team ingMode            : SwitchIndependent
    LoadBalancingAlgorithm : Dynamic
    Status                 : Up
    
    Name                   : bond_bond-backup
    Members                : {ens2f0, ens4f1}
    TeamNics               : bond-backup
    Team ingMode            : SwitchIndependent
    LoadBalancingAlgorithm : Dynamic
    Status                 : Up
    
    PS C:\> Get-NetAdapter
    
    Name                      InterfaceDescription                    ifIndex Status       MacAddress             LinkSpeed
    ----                      --------------------                    ------- ------       ----------             ---------
    bond-srv.9                Microsoft Network Adapter Multiple...#4      30 Up           40-A6-B7-27-96-D5        50 Gbps
    bond-srv                  Microsoft Network Adapter Multiple...#3      19 Up           40-A6-B7-27-96-D5        50 Gbps
    bond-iscsi                Microsoft Network Adapter Multiple...#2      18 Up           40-A6-B7-27-96-D4        50 Gbps
    bond-backup               Microsoft Network Adapter Multiplexo... 22 Up           68-05-CA-C9-EB-88        20 Gbps
    eno2                      Intel(R) Ethernet Connection X722 fo... 12 Disabled     38-68-DD-36-A0-59         1 Gbps
    ens3f0                    Intel(R) Ethernet Network Adapter XX... 11 Up           40-A6-B7-27-96-D4        25 Gbps
    ………………………………………….. Omitted
    PS C:\> Get-NetLbfoTeam
    
    Name                   : bond_bond-srv
    Members                : {ens6f1, ens3f1}
    TeamNics               : {bond-srv, bond-srv.9}
    Team ingMode            : SwitchIndependent
    LoadBalancingAlgorithm : Dynamic
    Status                 : Up
    
    Name                   : bond_bond-iscsi
    Members                : {ens6f0, ens3f0}
    TeamNics               : bond-iscsi
    Team ingMode            : SwitchIndependent
    LoadBalancingAlgorithm : Dynamic
    Status                 : Up
    
    Name                   : bond_bond-backup
    Members                : {ens2f0, ens4f1}
    TeamNics               : bond-backup
    Team ingMode            : SwitchIndependent
    LoadBalancingAlgorithm : Dynamic
    Status                 : Up
    
    PS C:\> Get-NetAdapter
    
    Name                      InterfaceDescription                    ifIndex Status       MacAddress             LinkSpeed
    ----                      --------------------                    ------- ------       ----------             ---------
    bond-srv.9                Microsoft Network Adapter Multiple...#4      30 Up           40-A6-B7-27-96-D5        50 Gbps
    bond-srv                  Microsoft Network Adapter Multiple...#3      19 Up           40-A6-B7-27-96-D5        50 Gbps
    bond-iscsi                Microsoft Network Adapter Multiple...#2      18 Up           40-A6-B7-27-96-D4        50 Gbps
    bond-backup               Microsoft Network Adapter Multiplexo... 22 Up           68-05-CA-C9-EB-88        20 Gbps
    eno2                      Intel(R) Ethernet Connection X722 fo... 12 Disabled     38-68-DD-36-A0-59         1 Gbps
    ens3f0                    Intel(R) Ethernet Network Adapter XX... 11 Up           40-A6-B7-27-96-D4        25 Gbps
    ………………………………………….. Omitted
    code block. Run Get-NetLbfoTeam

  6. Delete the Vlan information of the existing IP from the server.

    • Enter the verified ID in place of 30 in the following example.
      Color mode
      PS C:\> Remove-NetLbfoTeamNic -Team bond_bond-srv -VlanID 30 -Confirm:$false
      
      PS C:\> Get-NetAdapter
      
      Name                      InterfaceDescription                    ifIndex Status       MacAddress             LinkSpeed
      ----                      --------------------                    ------- ------       ----------             ---------
      bond-srv                  Microsoft Network Adapter Multiple...#3      19 Up           40-A6-B7-27-96-D5        50 Gbps
      bond-iscsi                Microsoft Network Adapter Multiple...#2      18 Up           40-A6-B7-27-96-D4        50 Gbps
      bond-backup               Microsoft Network Adapter Multiplexo... 22 Up           68-05-CA-C9-EB-88        20 Gbps
      eno2                      Intel(R) Ethernet Connection X722 fo... 12 Disabled     38-68-DD-36-A0-59         1 Gbps
      ens3f0                    Intel(R) Ethernet Network Adapter XX... 11 Up           40-A6-B7-27-96-D4        25 Gbps
      ………………………………………….. Omitted
      PS C:\> Remove-NetLbfoTeamNic -Team bond_bond-srv -VlanID 30 -Confirm:$false
      
      PS C:\> Get-NetAdapter
      
      Name                      InterfaceDescription                    ifIndex Status       MacAddress             LinkSpeed
      ----                      --------------------                    ------- ------       ----------             ---------
      bond-srv                  Microsoft Network Adapter Multiple...#3      19 Up           40-A6-B7-27-96-D5        50 Gbps
      bond-iscsi                Microsoft Network Adapter Multiple...#2      18 Up           40-A6-B7-27-96-D4        50 Gbps
      bond-backup               Microsoft Network Adapter Multiplexo... 22 Up           68-05-CA-C9-EB-88        20 Gbps
      eno2                      Intel(R) Ethernet Connection X722 fo... 12 Disabled     38-68-DD-36-A0-59         1 Gbps
      ens3f0                    Intel(R) Ethernet Network Adapter XX... 11 Up           40-A6-B7-27-96-D4        25 Gbps
      ………………………………………….. Omitted
      code block. Run Get-NetLbfoTeam
  7. After all tasks are completed, restart the server and then perform a final check.

    Reference
    It is recommended to perform the final check after restarting the server.

  8. If the final inspection results show no issues, select the work‑completion checkbox for Step 3 in the IP Change popup window.

Setting up RHEL Repository

The Samsung Cloud Platform Console provides the SCP RHEL Repository to support user environments where external access is restricted, such as VPC Private Subnet.
You can use the RHEL Repository to install and download the same packages as the official RHEL Repository.

To configure the RHEL Repository, refer to the RHEL Repository 설정 가이드.

안내
  • If a user created RHEL before August 2025 through the Samsung Cloud Platform Console, they must modify the RHEL Repository settings.
  • Since the SCP RHEL Repository synchronizes with each region’s local repository according to an internal schedule, it is recommended to switch to an external public mirror site to apply the latest patches quickly.
  • Samsung Cloud Platform provides the latest version repository for the given major version.

4.2.1 - Install ServiceWatch Agent

Users can install the ServiceWatch Agent on a Bare Metal Server to collect custom metrics and logs.

Reference
Collecting custom metrics/logs via the ServiceWatch Agent is currently available only on Samsung Cloud Platform For Enterprise. It will also be available in other offerings in the future.
Caution
Collecting metrics through the ServiceWatch Agent is classified as custom metrics and, unlike the default collected metrics, incurs charges; therefore, it is recommended to remove or disable unnecessary metric collection settings.

ServiceWatch Agent

On a Bare Metal Server, the agents that need to be installed to collect ServiceWatch’s custom metrics and logs can be divided into two main types. It is a Prometheus Exporter and Open Telemetry Collector.

CategoryDetailed description
Prometheus ExporterProvide metrics of a specific application or service in a format that Prometheus can scrape
  • To collect OS metrics from servers, you can use Node Exporter for Linux servers and Windows Exporter for Windows servers, depending on the OS type.
Open Telemetry CollectorActs as a centralized collector that gathers telemetry data such as metrics and logs from distributed systems, processes (filtering, sampling, etc.) it, and exports it to multiple backends (e.g., Prometheus, Jaeger, Elasticsearch, etc.)
  • Exports data to the ServiceWatch Gateway so that ServiceWatch can collect metric and log data.
Table. Explanation of Prometheus Exporter and Open Telemetry Collector
Reference
The ServiceWatch Agent guide can be used in the same way as Virtual Server. For more details, see Virtual Server > ServiceWatch Agent.

Pre-configuration for Using ServiceWatch Agent

To use the ServiceWatch Agent, please refer to Prerequisite Settings for ServiceWatch Agent and prepare the prerequisite settings.

4.2.2 - Setting up RHEL Repo and WKMS

The Samsung Cloud Platform Console provides the SCP RHEL Repository to support user environments where external access is restricted, such as VPC Private Subnet. You can use the SCP RHEL Repository to install and download the same packages as the official RHEL Repository.

guide
  • If a user created RHEL before August 2025 through the Samsung Cloud Platform Console, they must modify the RHEL Repository settings.
  • Since the SCP RHEL Repository synchronizes with each Region Local Repository according to an internal schedule, it is recommended to switch to an external public mirror site to apply the latest patches quickly.
  • Samsung Cloud Platform provides the latest repository for the specified major version.

RHEL Repository Configuration Guide

When using RHEL, you can configure the SCP RHEL Repository to install and download the same packages as the official RHEL Repository. To set up the RHEL Repository, follow these steps.

  1. On the Virtual Server, as the OS root user, use the cat command to check the /etc/yum.repos.d/scp.rhel8.repo or /etc/yum.repos.d/scp.rhel9.repo configuration.

    Color mode
    cat /etc/yum.repos.d/scp.rhel8.repo
    cat /etc/yum.repos.d/scp.rhel8.repo
    Code block. Verify repo configuration (RHEL8)
    Color mode
    cat /etc/yum.repos.d/scp.rhel9.repo
    cat /etc/yum.repos.d/scp.rhel9.repo
    Code block. Verify repo configuration (RHEL9)

    • When checking the configuration file, the following result is displayed.
      Color mode
      [rhel-8-baseos]
      name=rhel-8-baseos
      gpgcheck=0
      enabled=1
      baseurl=http://scp-rhel8-ip/rhel/8/baseos
      [rhel-8-baseos-debug]
      name=rhel-8-baseos-debug
      gpgcheck=0
      enabled=1
      baseurl=http://scp-rhel8-ip/rhel/8/baseos-debug
      [rhel-8-appstream]
      name=rhel-8-appstream
      gpgcheck=0
      enabled=1
      baseurl=http://scp-rhel8-ip/rhel/8/appstream
      [rhel-8-baseos]
      name=rhel-8-baseos
      gpgcheck=0
      enabled=1
      baseurl=http://scp-rhel8-ip/rhel/8/baseos
      [rhel-8-baseos-debug]
      name=rhel-8-baseos-debug
      gpgcheck=0
      enabled=1
      baseurl=http://scp-rhel8-ip/rhel/8/baseos-debug
      [rhel-8-appstream]
      name=rhel-8-appstream
      gpgcheck=0
      enabled=1
      baseurl=http://scp-rhel8-ip/rhel/8/appstream
      Code block. Check repo configuration (RHEL8)
      Color mode
      [rhel-9-for-x86_64-baseos-rpms]
      name=rhel-9-for-x86_64-baseos-rpms
      baseurl=http://scp-rhel9-ip/rhel/$releasever/x86_64/baseos
      gpgcheck=0
      enabled=1
      [rhel-9-for-x86_64-appstream-rpms]
      name=rhel-9-for-x86_64-appstream-rpms
      baseurl=http://scp-rhel9-ip/rhel/$releasever/x86_64/appstream
      gpgcheck=0
      enabled=1
      [codeready-builder-for-rhel-9-x86_64-rpms]
      name=codeready-builder-for-rhel-9-x86_64-rpms
      baseurl=http://scp-rhel9-ip/rhel/$releasever/x86_64/codeready-builder
      gpgcheck=0
      enabled=1
      [rhel-9-for-x86_64-highavailability-rpms]
      name=rhel-9-for-x86_64-highavailability-rpms
      baseurl=http://scp-rhel9-ip/rhel/$releasever/x86_64/ha
      gpgcheck=0
      enabled=1
      [rhel-9-for-x86_64-supplementary-rpms]
      name=rhel-9-for-x86_64-supplementary-rpms
      baseurl=http://scp-rhel9-ip/rhel/$releasever/x86_64/supplementary
      gpgcheck=0
      enabled=1
      [rhel-9-for-x86_64-baseos-rpms]
      name=rhel-9-for-x86_64-baseos-rpms
      baseurl=http://scp-rhel9-ip/rhel/$releasever/x86_64/baseos
      gpgcheck=0
      enabled=1
      [rhel-9-for-x86_64-appstream-rpms]
      name=rhel-9-for-x86_64-appstream-rpms
      baseurl=http://scp-rhel9-ip/rhel/$releasever/x86_64/appstream
      gpgcheck=0
      enabled=1
      [codeready-builder-for-rhel-9-x86_64-rpms]
      name=codeready-builder-for-rhel-9-x86_64-rpms
      baseurl=http://scp-rhel9-ip/rhel/$releasever/x86_64/codeready-builder
      gpgcheck=0
      enabled=1
      [rhel-9-for-x86_64-highavailability-rpms]
      name=rhel-9-for-x86_64-highavailability-rpms
      baseurl=http://scp-rhel9-ip/rhel/$releasever/x86_64/ha
      gpgcheck=0
      enabled=1
      [rhel-9-for-x86_64-supplementary-rpms]
      name=rhel-9-for-x86_64-supplementary-rpms
      baseurl=http://scp-rhel9-ip/rhel/$releasever/x86_64/supplementary
      gpgcheck=0
      enabled=1
      Code block. Verify repo configuration (RHEL9)
  2. Use a text editor (e.g., vim) to open the /etc/hosts file.

  3. Modify the /etc/hosts file with the following content and save it.

    Color mode
    198.19.2.13 scp-rhel8-ip scp-rhel9-ip scp-rhel-ip
    198.19.2.13 scp-rhel8-ip scp-rhel9-ip scp-rhel-ip
    code block. Change /etc/hosts file settings

  4. Use the yum command to verify the RHEL Repository connection configured on the server.

    Color mode
    yum repolist –v
    yum repolist –v
    Code block. Verify repository connection settings

    • If the RHEL repository is successfully connected, you can view the repository list.
      Color mode
      Repo-id            : rhel-8-appstream
      Repo-name          : rhel-8-appstream
      Repo-revision      : 1718903734
      Repo-updated       : Fri 21 Jun 2024 02:15:34 AM KST
      Repo-pkgs          : 38,260
      Repo-available-pkgs: 25,799
      Repo-size          : 122 G
      Repo-baseurl       : http://scp-rhel8-ip/rhel/8/appstream
      Repo-expire        : 172,800 second(s) (last: Thu 08 Aug 2024 07:27:57 AM KST)
      Repo-filename      : /etc/yum.repos.d/scp.rhel8.repo
      
      Repo-id            : rhel-8-baseos
      Repo-name          : rhel-8-baseos
      Repo-revision      : 1718029433
      Repo-updated       : Mon 10 Jun 2024 11:23:52 PM KST
      Repo-pkgs          : 17,487
      Repo-available-pkgs: 17,487
      Repo-size          : 32 G
      Repo-baseurl       : http://scp-rhel8-ip/rhel/8/baseos
      Repo-expire        : 172,800 second(s) (last: Thu 08 Aug 2024 07:27:57 AM KST)
      Repo-filename      : /etc/yum.repos.d/scp.rhel8.repo
      
      Repo-id            : rhel-8-baseos-debug
      Repo-name          : rhel-8-baseos-debug
      Repo-revision      : 1717662461
      Repo-updated       : Thu 06 Jun 2024 05:27:41 PM KST
      Repo-pkgs          : 17,078
      Repo-available-pkgs: 17,078
      Repo-size          : 100 G
      Repo-baseurl       : http://scp-rhel8-ip/rhel/8/baseos-debug
      Repo-expire        : 172,800 second(s) (last: Thu 08 Aug 2024 07:27:57 AM KST)
      Repo-filename      : /etc/yum.repos.d/scp.rhel8.repo
      Repo-id            : rhel-8-appstream
      Repo-name          : rhel-8-appstream
      Repo-revision      : 1718903734
      Repo-updated       : Fri 21 Jun 2024 02:15:34 AM KST
      Repo-pkgs          : 38,260
      Repo-available-pkgs: 25,799
      Repo-size          : 122 G
      Repo-baseurl       : http://scp-rhel8-ip/rhel/8/appstream
      Repo-expire        : 172,800 second(s) (last: Thu 08 Aug 2024 07:27:57 AM KST)
      Repo-filename      : /etc/yum.repos.d/scp.rhel8.repo
      
      Repo-id            : rhel-8-baseos
      Repo-name          : rhel-8-baseos
      Repo-revision      : 1718029433
      Repo-updated       : Mon 10 Jun 2024 11:23:52 PM KST
      Repo-pkgs          : 17,487
      Repo-available-pkgs: 17,487
      Repo-size          : 32 G
      Repo-baseurl       : http://scp-rhel8-ip/rhel/8/baseos
      Repo-expire        : 172,800 second(s) (last: Thu 08 Aug 2024 07:27:57 AM KST)
      Repo-filename      : /etc/yum.repos.d/scp.rhel8.repo
      
      Repo-id            : rhel-8-baseos-debug
      Repo-name          : rhel-8-baseos-debug
      Repo-revision      : 1717662461
      Repo-updated       : Thu 06 Jun 2024 05:27:41 PM KST
      Repo-pkgs          : 17,078
      Repo-available-pkgs: 17,078
      Repo-size          : 100 G
      Repo-baseurl       : http://scp-rhel8-ip/rhel/8/baseos-debug
      Repo-expire        : 172,800 second(s) (last: Thu 08 Aug 2024 07:27:57 AM KST)
      Repo-filename      : /etc/yum.repos.d/scp.rhel8.repo
      Code block. Check repository list

Windows Key Management Service Configuration Guide

When using Windows Server on Samsung Cloud Platform, you can authenticate genuine licenses by utilizing the Key Management Service provided by Samsung Cloud Platform. To authenticate genuine copies using the Key Management Service, follow these steps.

  1. Right-click the Windows Start icon, then run cmd from Windows PowerShell (Administrator) or the Windows Run menu.

  2. Run the following command in Windows PowerShell(Administrator) or cmd to register the KMS Server.

    Color mode
    slmgr /skms 198.19.2.23:1688
    slmgr /skms 198.19.2.23:1688
    code block. WKMS configuration

  3. Run the KMS Server registration command, verify the notification popup indicating successful registration, and then click the OK button.

    Figure

  4. Run the following command in Windows PowerShell(Administrator) or cmd to activate Windows.

    Color mode
    slmgr /ato
    slmgr /ato
    Code block. Windows Server activation settings

  5. After confirming the notification popup indicating that activation was successful, click the OK button.

    Figure

  6. In Windows PowerShell(Administrator) or cmd, run the following command to verify that the product is properly activated.

    Color mode
    slmgr /dlv
    slmgr /dlv
    Code block. Verify Windows Server activation

  7. After confirming the notification popup indicating that product activation was successful, click the OK button.

    Figure

4.3 - API Reference

API Reference

4.4 - CLI Reference

CLI Reference

4.5 - Release Note

Bare Metal Server

2026.07.16
FEATURE CPU configuration option support and other feature changes
  • CPU configuration options are supported.
    • Through the CPU setting options, you can activate only the desired number of CPU cores and adjust the number of threads per core.
    • Even if you change the number of cores, the usage fee remains the same.
  • When creating a Bare Metal Server, up to 20 can be created simultaneously.
    • When creating two or more servers, you can automatically generate names in the format server name + creation number.

2026.03.19
FEATURE Add new server type
  • The 4th generation BM based on the Intel 6th generation (Granite Rapids) Processor has been released.
    • Bare Metal Server s4 and h4 server types have been added.
    • Detailed information about the s4/h4 server type can be found in the s3/h3 server type.
2025.10.23
FEATURE Add Local disk partition feature
  • Provides local disk partition functionality
    • Now you can create and use up to 10 Local disk partitions.
2025.07.01
FEATURE Add new features and provide additional OS Image
  • You can terminate multiple resources simultaneously from the Bare Metal Server list.
  • You can change the IP of a regular Subnet.
  • OS Image has been added.
    • RHEL 8.10, Ubuntu 24.04
2025.02.27
FEATURE Placement Group functionality and addition of OS Image, server type
  • Add Bare Metal Server feature
    • Distribute servers belonging to the same Placement group across different racks.
    • Additional OS images provided (RHEL 9.4, Rocky Linux 8.6, Rocky Linux 9.4)
    • Add 3rd‑generation (s3/h3) server type based on Intel 4th‑generation (Sapphire Rapids) Processor. * For detailed information, refer to Bare Metal Server 서버 타입.
  • Samsung Cloud Platform Common Feature Change
    • Account, IAM, and Service Home, tags, etc., have been updated to reflect common CX changes.
2024.10.01
NEW Bare Metal Server Service Official Version Release
  • We have officially launched the Bare Metal Server service.
  • We have launched a Bare Metal Server service that lets customers use a physical server without virtualization, exclusively for themselves.

5 - Multi-node GPU Cluster

5.1 - Overview

Service Overview

Multi-node GPU Cluster is a service that provides physical GPU servers without virtualization for large-scale high-performance AI computation. You can use two or more Bare Metal Servers equipped with GPUs to cluster multiple GPUs, and conveniently operate GPU servers in conjunction with Samsung Cloud Platform’s high‑performance storage and networking services.

Provided Features

The Multi-node GPU Cluster provides the following features.

  • Auto Provisioning and Management: Through the web-based Console, you can easily provision servers of the standard GPU Bare Metal model equipped with 8 GPUs and manage resources and costs.
  • Network Connection: You can cluster multiple GPUs on two or more Bare Metal Servers via high‑speed interconnects, and by configuring a GPU Direct RDMA (Remote Direct Memory Access) environment, you can directly process data I/O between GPU memories, enabling high‑speed AI/Machine Learning computation.
  • Storage Connection: Provides various additional attached storage besides the OS disk. * High-performance SSD NAS File Storage, Block Storage, and Object Storage directly integrated with a high-speed network can also be used together.
  • Network Configuration Management: The server’s subnet/IP can be easily changed from the values set at initial creation. * NAT IP provides a management feature that allows you to enable or disable it as needed.
  • Monitoring: You can view monitoring information for computing resources such as CPU, GPU, Memory, and Disk through Cloud Monitoring. * To use the Cloud Monitoring service of a Multi-node GPU Cluster, you need to install the Agent. * Please install the Agent to ensure stable service. * For more details, please refer to Multi-node GPU Cluster Monitoring Metrics.
  • Terraform Provision: Provides an IaC environment via Terraform.

Component

Multi-node GPU Cluster provides GPUs as a Bare Metal Sever type with standard images and server types. NVSwitch and NVLink are provided.

Specifications by GPU Type

GPU (Graphic Processing Unit) is specialized for parallel operations that process large amounts of data quickly, enabling large-scale parallel computation in fields such as artificial intelligence (AI) and data analysis.

The following are the specifications of GPU types offered by the Multi-node GPU Cluster service.

CategoryH100 TypeB300 Type
GPU ArchitectureNVIDIA HopperNVIDIA Blackwell Ultra
GPU Memory80 GiB268 GiB
GPU Transistors80 billion 4N TSMC208 billion 4NP TSMC
FP16 Tensor Core (Dense)989 TFLOPs2.25 PFLOPs
FP8 Tensor Core (Dense)1979 TFLOPs4.5 PFLOPs
FP4 Tensor Core (Dense)Not supported13.5 PFLOPs
GPU Memory Bandwidth3,352 GB/s HBM38 TB/s HBM3e
NVLink performanceNVLink 4NVLink 5
NVLink Signaling Rate25 GB/s (x18)50 GB/s (x18)
NVSwitch GPU-to-GPU bandwidth900 GB/s1.8 TB/s
Total NVSwitch aggregate bandwidth7.2 TB/s14.4 TB/s
Table. GPU Type specifications

OS and GPU driver version

The operating systems (OS) supported by the Multi-node GPU Cluster are as follows.

OSOS versionGPU driver version
Ubuntu22.04535.86.10, 535.183.06
Ubuntu24.04580.105.08
Table. Multi-node GPU Cluster OS and GPU driver version

Server type

The format of server types provided by the Multi-node GPU Cluster is as follows.

  • Example: when the server type is g2c96h8_metal
CategoryexampleDetailed description
Server generationg2Provided server generation
  • g2: g means GPU server, and 2 means generation
CPUc96Number of cores
  • c96: Allocated cores are physical cores
GPUh8GPU type and quantity
  • h8: h means GPU type, and 8 means GPU quantity
Table. Multi-node GPU Cluster server type format
Reference
For detailed information about the server types provided by Multi-node GPU Cluster, refer to Multi-node GPU Cluster Server Types.

Preceding Service

This is a list of services that must be pre-configured before creating the service. Please refer to the guide provided for each service and prepare in advance.

Service CategoryserviceDetailed description
NetworkingVPCA service that provides an isolated virtual network in a cloud environment
Table. Multi-node GPU Cluster preliminary service

5.1.1 - Server type

Multi-node GPU Cluster server type

Multi-node GPU Cluster is categorized based on the GPU Type it provides, and the GPU used in a Multi-node GPU Cluster is determined by the server type selected when creating a GPU Node. Select the server type based on the specifications of the application you want to run on a multi-node GPU cluster.

The server types supported by the Multi-node GPU Cluster are as follows.

  • Example: when the server type is g2c96h8_metal.
    CategoryexampleDetailed description
    Server generationg2Provided server generation
    • g2
      • g means GPU server specification
      • 2 means generation
    CPUc96Number of cores
    • c96: Allocated cores are physical cores
    GPUh8GPU type and quantity
    • h8: h means GPU type, and 8 means GPU quantity
    Table. Multi-node GPU Cluster server type format

g2 server type

The g2 server type is a GPU Bare Metal Server that uses NVIDIA H100 SXM GPUs, suitable for large-scale high-performance AI computation.

  • 8 NVIDIA Hopper Architecture-based H100 GPUs provided
  • Provides 1,979 TFLOPS FP8 Tensor Core performance per GPU, 989 TFLOPS FP16 Tensor Core performance.
  • Supports up to 96 vCPUs and 2,048 GB of memory
  • Supports up to 1,600 Gb/s NVIDIA InfiniBand RDMA network.
  • Service network up to 100 Gbps
  • 900 GB/s GPU P2P communication via NVSwitch within a node
Server typeGPUGPU MemoryCPU(Core)MemoryDiskGPU P2P
g2c96h8_metalH100640 GiB96 vCore2 TBSSD (OS) 960 GB * 2, NVMeSSD 3.84 TB * 4900 GB/s NVSwitch
Table. Multi-node GPU Cluster server type specifications > H100 server type

g3 server type

The g3 server type is a GPU Bare Metal Server that uses NVIDIA B300 SXM GPUs, suitable not only for large-scale high-performance AI computation but also for LLM inference and AI deployment for generative AI.

  • 8 NVIDIA Blackwell Ultra Architecture-based B300 GPUs provided
  • Provides 13.5 PFLOPS FP4 Tensor Core and 4.5 PFLOPS FP8 Tensor Core performance per GPU.
  • Supports up to 128 vCPUs and 4,096 GB of memory
  • Supports up to 6,400 Gb/s NVIDIA InfiniBand RDMA network
  • Service network up to 100 Gbps
  • 1.8 TB/s GPU P2P communication via NVSwitch within a node
Server typeGPUGPU MemoryCPU(Core)MemoryDiskGPU P2P
g3c128b8_metalB3002.1 TiB128 vCore4 TBSSD (OS) 960 GB * 2, NVMeSSD 3.84 TB * 41.8 TB/s NVSwitch
Table. Multi-node GPU Cluster server type specifications > B300 server type

5.1.2 - Monitoring Metrics

Cloud Monitoring service termination notice

According to Samsung Cloud Platform’s policy, the Cloud Monitoring service is scheduled to be discontinued in September 2026.
Accordingly, from after the September 2026 release, resource monitoring of the Samsung Cloud Platform via Cloud Monitoring will no longer be possible.

With a new alternative service, you can continuously perform resource monitoring by leveraging ServiceWatch released in October 2025.
ServiceWatch provides more modern and powerful features, replacing Cloud Monitoring to deliver a seamless monitoring environment.

If you are collecting metrics and logs through the Cloud Monitoring Agent, you need to switch to the ServiceWatch Agent.

For more details about ServiceWatch, please refer to ServiceWatch Overview.
Detailed information about ServiceWatch Agent: please refer to the ServiceWatch Agent

Multi-node GPU Cluster Monitoring Metrics

The table below shows the monitoring metrics of a Multi-node GPU Cluster that can be viewed through Cloud Monitoring.

Guide
In a Multi-node GPU Cluster, users must install the Agent themselves via the guide to view monitoring metrics. Before using the stable service, please be sure to install the Agent. For instructions on installing the Agent and detailed usage of Cloud Monitoring, refer to the Cloud Monitoring guide.

Multi-node GPU Cluster [Cluster]

Performance itemsDetailed descriptionunit
Memory Total [Basic]bytes of usable memorybytes
Memory Used [Basic]Current memory usage in bytesbytes
Memory Swap In [Basic]bytes of the replaced memorybytes
Memory Swap Out [Basic]bytes of the replaced memorybytes
Memory Free [Basic]bytes of unused memorybytes
Disk Read Bytes [Basic]Read bytesbytes
Disk Read Requests [Basic]Number of read requestscnt
Disk Write Bytes [Basic]write bytesbytes
Disk Write Requests [Basic]Number of write requestscnt
CPU Usage [Basic]Average system CPU usage over 1 minute%
Instance State [Basic]Instance statusstate
Network In Bytes [Basic]Received bytesbytes
Network In Dropped [Basic]Incoming packet dropcnt
Network In Packets [Basic]Number of received packetscnt
Network Out Bytes [Basic]sent bytesbytes
Network Out Dropped [Basic]Transmit packet dropcnt
Network Out Packets [Basic]Number of transmitted packetscnt
Table. Multi-node GPU Cluster [Cluster] Monitoring Metrics (default)
Performance itemsDetailed descriptionunit
Cluster GPU CountGPU Count SUM in Cluster
  • Sum of GPU Count for nodes in the cluster: calculate the total GPU Count of each node within the same GPU CLUSTER
cnt
Cluster GPU Count In UseNumber of GPUs being used by jobs within the cluster
  • Number of GPUs used by processes within the cluster: sum of GPUs occupied by processes, parsed from the ‘Processes:’ section at the bottom of nvidia-smi output of nodes in the same GPU cluster
cnt
Cluster GPU UsageGPU Utilization AVG within the cluster
  • Cluster node GPU utilization average value: calculate the average of each node’s GPU utilization values within the same GPU cluster
%
Cluster GPU Memory Usage [Avg]Cluster GPU Memory Utilization AVG
  • Average Memory utilization of nodes within the cluster: calculate the average of each node’s Memory utilization values among nodes in the same GPU cluster
%
Table. Multi-node GPU Cluster [Cluster] Additional monitoring metrics (Agent installation required)

Multi-node GPU Cluster [Node]

Performance itemsDetailed descriptionunit
Memory Total [Basic]bytes of usable memorybytes
Memory Used [Basic]Current memory usage in bytesbytes
Memory Swap In [Basic]bytes of the replaced memorybytes
Memory Swap Out [Basic]bytes of the replaced memorybytes
Memory Free [Basic]bytes of unused memorybytes
Disk Read Bytes [Basic]Read bytesbytes
Disk Read Requests [Basic]Number of read requestscnt
Disk Write Bytes [Basic]write bytesbytes
Disk Write Requests [Basic]Number of write requestscnt
CPU Usage [Basic]Average system CPU usage over 1 minute%
Instance State [Basic]Instance statusstate
Network In Bytes [Basic]Received bytesbytes
Network In Dropped [Basic]Incoming packet dropcnt
Network In Packets [Basic]Number of received packetscnt
Network Out Bytes [Basic]sent bytesbytes
Network Out Dropped [Basic]Transmit packet dropcnt
Network Out Packets [Basic]Number of transmitted packetscnt
Table. Multi-node GPU Cluster [Node] Monitoring Metrics (provided by default)
Performance itemsDetailed descriptionunit
GPU CountNumber of GPUscnt
GPU TemperatureGPU temperature
GPU Usageutilization%
GPU Usage [Avg]Overall average GPU utilization (%)%
GPU Power CapMaximum power capacity of the GPUW
GPU Power UsageCurrent GPU power usageW
GPU Memory Usage [Avg]GPU Memory Uti. AVG%
GPU Count in useNumber of GPUs in use by jobs on the nodecnt
Execution Status for nvidia-smiResult of running the nvidia-smi commandstatus
Core Usage [IO Wait]Ratio of CPU time spent in wait state (disk wait)%
Core Usage [System]Proportion of CPU time spent in kernel space%
Core Usage [User]Proportion of CPU time spent in user space%
CPU CoresThe number of CPU cores on the host. The maximum value of the unnormalized ratio is 100%* of a core. The unnormalized ratio already incorporates this value, and its maximum is 100%* of a core.cnt
CPU Usage [Active]Percentage of CPU time used excluding Idle and IOWait states (when all 4 cores are used at 100%: 400%)%
CPU Usage [Idle]It is the proportion of CPU time spent in idle state.%
CPU Usage [IO Wait]This is the proportion of CPU time spent in a waiting state (disk wait).%
CPU Usage [System]Percentage of CPU time used by the kernel (when all 4 cores are used at 100%: 400%)%
CPU Usage [User]Percentage of CPU time used in user space. (If all 4 cores are used at 100%, it is 400%)%
CPU Usage/Core [Active]Percentage of CPU time used excluding Idle and IOWait states (value normalized by the number of cores; 100% when all four cores are fully utilized)%
CPU Usage/Core [Idle]It is the proportion of CPU time spent in idle state.%
CPU Usage/Core [IO Wait]This is the proportion of CPU time spent in a waiting state (disk wait).%
CPU Usage/Core [System]Percentage of CPU time used by the kernel (value normalized by the number of cores; 100% when all 4 cores are fully utilized)%
CPU Usage/Core [User]Percentage of CPU time used in user space. (Value normalized by the number of cores; 100% when all 4 cores are fully utilized)%
Disk CPU Usage [IO Request]It is the proportion of CPU time during which I/O requests for the device were executed (device bandwidth utilization). If this value approaches 100%, the device becomes saturated.%
Disk Queue Size [Avg]The average queue length of requests executed for the device.num
Disk Read BytesThe number of bytes read per second from the device.bytes
Disk Read Bytes [Delta Avg]Average of system.diskio.read.bytes_delta for individual disksbytes
Disk Read Bytes [Delta Max]Maximum system.diskio.read.bytes_delta of individual disksbytes
Disk Read Bytes [Delta Min]Minimum system.diskio.read.bytes_delta of individual disksbytes
Disk Read Bytes [Delta Sum]Sum of the system.diskio.read.bytes_delta of individual disksbytes
Disk Read Bytes [Delta]Delta of the system.diskio.read.bytes value for each diskbytes
Disk Read Bytes [Success]Total number of bytes successfully read. On Linux, assuming a sector size of 512, it is the number of sectors read multiplied by 512.bytes
Disk Read RequestsNumber of read requests to the disk device per secondcnt
Disk Read Requests [Delta Avg]Average of the system.diskio.read.count_delta for individual diskscnt
Disk Read Requests [Delta Max]Maximum system.diskio.read.count_delta for individual diskscnt
Disk Read Requests [Delta Min]Minimum of system.diskio.read.count_delta for individual diskscnt
Disk Read Requests [Delta Sum]Sum of system.diskio.read.count_delta of individual diskscnt
Disk Read Requests [Success Delta]Delta of system.diskio.read.count for each diskcnt
Disk Read Requests [Success]Total number of successful readscnt
Disk Request Size [Avg]Average size of requests executed on the device (unit: sectors).num
Disk Service Time [Avg]Average service time (ms) of input requests executed on the device.ms
Disk Wait Time [Avg]Average time taken for requests executed on the supported device.ms
Disk Wait Time [Read]Average disk wait timems
Disk Wait Time [Write]Average disk wait timems
Disk Write Bytes [Delta Avg]Average of system.diskio.write.bytes_delta for each diskbytes
Disk Write Bytes [Delta Max]Maximum system.diskio.write.bytes_delta of individual disksbytes
Disk Write Bytes [Delta Min]Minimum of system.diskio.write.bytes_delta for individual disksbytes
Disk Write Bytes [Delta Sum]Sum of the system.diskio.write.bytes_delta of individual disksbytes
Disk Write Bytes [Delta]Delta of the system.diskio.write.bytes value for each diskbytes
Disk Write Bytes [Success]Total number of bytes successfully written. On Linux, assuming a sector size of 512, it is the number of sectors written multiplied by 512.bytes
Disk Write RequestsNumber of write requests to the disk device per secondcnt
Disk Write Requests [Delta Avg]Average of system.diskio.write.count_delta for individual diskscnt
Disk Write Requests [Delta Max]Maximum system.diskio.write.count_delta for individual diskscnt
Disk Write Requests [Delta Min]Minimum of system.diskio.write.count_delta for individual diskscnt
Disk Write Requests [Delta Sum]Sum of the system.diskio.write.count_delta of individual diskscnt
Disk Write Requests [Success Delta]Delta of system.diskio.write.count for each diskcnt
Disk Write Requests [Success]Total number of successful writescnt
Disk Writes BytesIt is the number of bytes per second written to the device.bytes
Filesystem Hang Checkfilesystem (local/NFS) hang check (normal:1, abnormal:0)status
Filesystem NodesIt is the total number of file nodes in the file system.cnt
Filesystem Nodes [Free]It is the total number of available file nodes in the file system.cnt
Filesystem Size [Available]Disk space (bytes) that unauthorized users can use.bytes
Filesystem Size [Free]Available disk space (bytes)bytes
Filesystem Size [Total]Total disk space (bytes)bytes
Filesystem UsageUsed disk space percentage%
Filesystem Usage [Avg]Average of individual filesystem.used.pct%
Filesystem Usage [Inode]inode usage%
Filesystem Usage [Max]Maximum among individual filesystem.used.pct%
Filesystem Usage [Min]minimum of individual filesystem.used.pct%
Filesystem Usage [Total]-%
Filesystem UsedUsed disk space (bytes)bytes
Filesystem Used [Inode]inode usagebytes
Memory FreeTotal amount of available memory (bytes). Memory used by system cache and buffers is not included (see system.memory.actual.free).bytes
Memory Free [Actual]Actual usable memory (bytes). The calculation method varies by OS: on Linux, it is MemAvailable from /proc/meminfo, or if meminfo cannot be used, it is calculated from available memory plus cache and buffers. On OSX, it is the sum of usable memory and inactive memory. On Windows, it corresponds to a value such as system.memory.free.bytes
Memory Free [Swap]Available swap memory.bytes
Memory Totaltotal memorybytes
Memory Total [Swap]Total swap memory.bytes
Memory UsagePercentage of used memory
  • ((Memory Total - Memory Free) / Memory Total) * 100
  • Memory Free: the amount of available memory currently free
%
Memory Usage [Actual]Percentage of memory actually used
  • ((Memory Total - Mememory Available) / Memory Total) * 100 or ((Memory Total - (Memmory Free + Buffers + Cached) / MemTotal) * 100
  • Memory Free: the amount of free memory currently available
  • Buffers: the amount of memory used for buffers
  • Cached: the amount of memory used for the page cache
%
Memory Usage [Cache Swap]Cached swap usage rate%
Memory Usage [Swap]Percentage of used swap memory%
Memory Usedused memorybytes
Memory Used [Actual]Actual used memory (bytes). The value obtained by subtracting used memory from total memory. Available memory is calculated differently for each OS (see system.actual.free).bytes
Memory Used [Swap]Used swap memory.bytes
CollisionsNetwork collisioncnt
Network In BytesNumber of received bytesbytes
Network In Bytes [Delta Avg]Average of system.network.in.bytes_delta for each networkbytes
Network In Bytes [Delta Max]Maximum of system.network.in.bytes_delta for each networkbytes
Network In Bytes [Delta Min]Minimum system.network.in.bytes_delta for each networkbytes
Network In Bytes [Delta Sum]Sum of system.network.in.bytes_delta for individual networksbytes
Network In Bytes [Delta]Delta of received byte countbytes
Network In DroppedNumber of deleted packets among incoming packetscnt
Network In ErrorsNumber of errors during receptioncnt
Network In PacketsNumber of received packetscnt
Network In Packets [Delta Avg]Average of system.network.in.packets_delta for individual networkscnt
Network In Packets [Delta Max]Maximum of system.network.in.packets_delta for each networkcnt
Network In Packets [Delta Min]Minimum of system.network.in.packets_delta for each networkcnt
Network In Packets [Delta Sum]Sum of system.network.in.packets_delta for individual networkscnt
Network In Packets [Delta]Delta of received packet countcnt
Network Out BytesNumber of transmitted bytesbytes
Network Out Bytes [Delta Avg]Average of system.network.out.bytes_delta for each networkbytes
Network Out Bytes [Delta Max]Maximum system.network.out.bytes_delta of individual networksbytes
Network Out Bytes [Delta Min]Minimum of system.network.out.bytes_delta for individual networksbytes
Network Out Bytes [Delta Sum]Sum of system.network.out.bytes_delta for individual networksbytes
Network Out Bytes [Delta]Delta of transmitted byte countbytes
Network Out DroppedNumber of deleted packets among outgoing packets. This value is not reported by the operating system, so it is always 0 on Darwin and BSD.cnt
Network Out ErrorsNumber of errors during transmissioncnt
Network Out PacketsNumber of transmitted packetscnt
Network Out Packets [Delta Avg]Average of system.network.out.packets_delta for each networkcnt
Network Out Packets [Delta Max]Maximum of system.network.out.packets_delta for each networkcnt
Network Out Packets [Delta Min]Minimum of system.network.out.packets_delta for each networkcnt
Network Out Packets [Delta Sum]Sum of system.network.out.packets_delta for individual networkscnt
Network Out Packets [Delta]Delta of transmitted packet countcnt
Open Connections [TCP]All open TCP connectionscnt
Open Connections [UDP]All open UDP connectionscnt
Port UsageAvailable port usage rate%
SYN Sent SocketsNumber of sockets in SYN_SENT state (when connecting from local to remote)cnt
Kernel PID Maxkernel.pid_max valuecnt
Kernel Thread Maxkernel.threads-max valuecnt
Process CPU UsagePercentage of CPU time consumed by the process since the last update. This value is similar to the %CPU value shown for the process by the top command on Unix systems.%
Process CPU Usage/CoreThe percentage of CPU time used by the process since the last event. Normalized by the number of cores, with values ranging from 0 to 100%.%
Process Memory UsageProportion of main memory (RAM) occupied by a process%
Process Memory UsedResident Set size. The amount of memory a process occupies in RAM. In Windows, the current working set size.bytes
Process PIDprocess pidPID
Process PPIDparent process PIDPID
Processes [Dead]Number of dead processescnt
Processes [Idle]Number of idle processescnt
Processes [Running]Number of running processescnt
Processes [Sleeping]Number of sleeping processescnt
Processes [Stopped]stopped processes countcnt
Processes [Total]Total number of processescnt
Processes [Unknown]Number of processes with an unknown or unsearchable statuscnt
Processes [Zombie]Number of zombie processescnt
Running Process Usageprocess usage%
Running ProcessesNumber of running processescnt
Running Thread UsageThread usage rate%
Running ThreadsTotal number of threads running in running processescnt
Instance StatusInstance statusstate
Context Switchescontext switch count (per second)cnt
Load/Core [1 min]The load over the last 1 minute divided by the number of corescnt
Load/Core [15 min]The load over the last 15 minutes divided by the number of corescnt
Load/Core [5 min]The load over the last 5 minutes divided by the number of corescnt
Multipaths [Active]External storage connection path status = active countcnt
Multipaths [Failed]External storage connection path status = failed countcnt
Multipaths [Faulty]External storage connection path status = faulty countcnt
NTP Offsetmeasured offset of the last sample (the time difference between the NTP server and the local environment)num
Run Queue LengthExecution queue lengthnum
UptimeOS uptime(uptime). (milliseconds)ms
Context SwitchiesCPU context switch count (per second)cnt
Disk Read Bytes [Sec]Number of bytes read from a Windows logical disk in 1 secondcnt
Disk Read Time [Avg]Average data read time (seconds)sec
Disk Transfer Time [Avg]Disk average wait timesec
Disk UsageDisk usage%
Disk Write Bytes [Sec]Number of bytes written in one second on a Windows logical diskcnt
Disk Write Time [Avg]Average data write time (seconds)sec
Pagingfile UsagePaging file usage%
Pool Used [Non Paged]Nonpaged Pool usage in kernel memorybytes
Pool Used [Paged]Paged Pool usage in kernel memorybytes
Process [Running]Number of currently running processescnt
Threads [Running]Number of currently running threadscnt
Threads [Waiting]Number of threads waiting for processor timecnt
Table. Multi-node GPU Cluster [Node] Additional monitoring metrics (Agent installation required)

5.2 - How-to guides

Users can create the service by entering the required information for the Multi-node GPU Cluster service and selecting detailed options through the Samsung Cloud Platform Console.

Multi-node GPU Cluster Getting Started

You can create and use a Multi-node GPU Cluster service from the Samsung Cloud Platform Console.

This service consists of a GPU Node and a Cluster Fabric service.

Create GPU Node

Follow the steps below to create a Multi-node GPU Cluster.

  1. All Services > Compute > Multi-node GPU Cluster Click the menu. 1. Go to the Service Home page of the Multi-node GPU Cluster.

  2. On the Service Home page, click the Create GPU Node button. 2. Navigate to the GPU Node Creation page.

  3. On the GPU Node creation page, enter the information required to create the service, and select detailed options.

    • Select the required information in the Image and Version Selection area.
      Category
      required status
      Detailed description
      imageRequiredSelect the type of image provided
      • Ubuntu
      Image versionRequiredSelect version of the selected image
      • Provide version list of the provided server image
      Table. GPU Node image and version selection options
    • Enter or select the required information in the Service Information Input area.
      Category
      required status
      Detailed description
      Number of serversRequiredNumber of GPU Node servers to create simultaneously
      • Only numeric input is allowed, and the minimum number of servers to create is 2.
      • When initially configuring, you must create at least 2, and expansions can be done one at a time.
      Service Type > Server TypeRequiredGPU Node server type
      • Select the desired CPU, Memory, GPU, and Disk specifications
      Service Type > Planned ComputeRequiredPlanned Compute가 설정된 자원 현황
      • In Use: Number of resources with Planned Compute that are currently in use
      • Configured: Number of resources with Planned Compute set
      • Coverage Preview: Amount applied per resource by Planned Compute
      Table. GPU Node Service Information Input Items
    • In the Required Information Input area, enter or select the required information.
      Category
      required status
      Detailed description
      Administrator accountRequiredSet the administrator account and password to be used when connecting to the server
      • Ubuntu OS is provided with a fixed root account
      Server name PrefixRequiredEnter a prefix to distinguish each GPU Node generated when the selected number of servers is 2 or more
      • Automatically generated in the form of user input value (prefix) + ‘-###
      • Must start with a lowercase English letter and be entered using lowercase letters, numbers, and special characters (-) within 3 to 11 characters
      • Must not end with a special character (-)
      Network SettingsRequiredSet the network where the GPU Node will be installed
      • VPC Name: Select a pre‑created VPC
      • General Subnet Name: Select a pre‑created standard Subnet
        • IP can be Auto‑generated or Manual Input; if Manual Input is selected, the user enters the IP directly
      • NAT: Available only when there is a single server and the VPC is attached to an Internet Gateway. Checking the option allows selection of a NAT IP. (Initially, only configurations with two or more servers can be created, so modify on the resource detail page)
      • NAT IP: Select a NAT IP
        • If no NAT IP is available to select, click the Create New button to generate a Public IP
        • Refresh button to view and select the created Public IP
        • Creating a Public IP incurs charges according to the Public IP pricing policy
      Table. GPU Node required information input items
    • In the Cluster Selection area, create or select a Cluster Fabric.
      Category
      required status
      Detailed description
      Cluster FabricRequiredConfiguration of GPU Node server groups that can apply GPU Direct RDMA together
      • Optimal GPU performance and speed can be secured only within the same Cluster Fabric
      • When creating a new Cluster Fabric, *New Input > select Node pool, then enter the name of the Cluster Fabric to be created
      • To add to an existing Cluster Fabric, Existing Input > select Node pool, then choose the previously created Cluster Fabric
      Table. GPU Node Cluster Fabric options
    • In the Additional Information Input area, enter or select the required information.
      Category
      required status
      Detailed description
      LockSelectionUsing a lock prevents actions caused by mistakes, such as terminating, starting, or stopping the server.
      Init ScriptSelectionScript to run at server startup
      • The Init Script must be selected differently depending on the image type
        • For Linux: Choose Shell Script or cloud-init
      tagSelectionAdd Tag
      • You can add up to 50 per resource
      • After clicking the Add Tag button, enter or select Key, Value values
      Table. GPU Node additional information input fields
  4. Summary Check the detailed information and estimated billing amount generated in the panel, and click the Create button.

  5. When the popup notifying creation opens, click the Confirm button.

    • When creation is complete, check the created resources on the GPU Node List page.
Caution
  • When creating a service, the GPU MIG/ECC settings are reset. * However, to apply the correct settings, perform an initial reboot, verify that the settings have been applied, and then use it.
  • For detailed information on resetting GPU MIG/ECC settings, refer to the GPU MIG/ECC 설정 초기화 점검 가이드.

Check GPU Node detailed information

The Multi-node GPU Cluster service can view and modify the full list of GPU Node resources and detailed information.

GPU Node Details page consists of Details, Tags, Job History tabs.

To view detailed information about the GPU Node, follow these steps.

  1. Click the All Services > Compute > Multi-node GPU Cluster > GPU Node menu. 1. Go to the Service Home page of the Multi-node GPU Cluster.

  2. On the Service Home page, click the GPU Node menu. 2. Go to the GPU Node list page.

    • Resource items other than the required columns can be added through the Settings button.
      Category
      required status
      Detailed description
      Resource IDSelectionUser-created GPU Node ID
      Cluster Fabric nameRequiredUser-created Cluster Fabric name
      Server nameRequiredUser-created GPU Node name
      Server typeRequiredServer type of GPU Node
      • The user can view the number of cores, memory capacity, and GPU type and count of the created resources
      imageRequiredUser-generated GPU Node image version
      IPRequiredIP of the GPU node created by the user
      statusRequiredStatus of the GPU Node created by the user
      Creation date and timeSelectionGPU Node creation timestamp
      Table. GPU Node resource list items
  3. GPU Node List page, click the resource to view detailed information. 3. Go to the GPU Node Details page.

    • GPU Server Details At the top of the page, status information and descriptions of additional features are displayed.
      CategoryDetailed description
      GPU Node statusStatus of the GPU Node created by the user
      • Creating: State while the server is being created
      • Running:: State when creation is complete and the server is available for use
      • Editing:: State while the IP is being changed
      • Unknown: Error state
      • Starting: State while the server is starting
      • Stopping: State while the server is stopping
      • Stopped: State when the server has stopped
      • Terminating: State while terminating
      • Terminated: State when termination is complete
      Server controlButton to change server status
      • Start: Start a stopped server
      • Stop: Stop a running server
      Service cancellationCancel service button
      Table. GPU Node status information and additional features

Detailed Information

On the GPU Node List page’s Details Tab, you can view the detailed information of the selected resource and edit the information if needed.

CategoryDetailed description
serviceservice name
Resource TypeResource Type
SRNUnique resource ID in Samsung Cloud Platform
  • In a GPU Node, it means the GPU Node SRN
Resource nameResource name
  • In the GPU Node service, it refers to the GPU Node name
Resource IDUnique resource ID in the service
ConstructorUser who created the service
Creation date and timeService creation date and time
ModifierUser who edited the service information
Modification date and timeDate and time the service information was modified
Server nameserver name
Node poolA collection of nodes that can be grouped into the same Cluster Fabric
Cluster Fabric nameUser-created Cluster Fabric name
Image/VersionServer OS image and version
Server typeCPU, memory, GPU, information display
Planned ComputeResource status with Planned Compute configured
LockIndicates whether Lock is enabled/disabled
  • When Lock is enabled, it prevents server termination/start/stop actions, avoiding accidental operations
  • If you need to change the Lock property value, click the Edit button to set it
NetworkGPU Node network information
  • VPC name, regular Subnet name, IP, Public NAT IP, and status
Block StorageBlock Storage information connected to the server
  • Volume name, disk type, capacity, status
Init ScriptView the Init Script content entered during server creation
Table. GPU Node detailed information tab items

note
If the VPC does not have an Internet Gateway attached, you cannot attach a Public NAT IP.

Tag

On the GPU Node List page’s Tag Tab, you can view the selected resource’s tag information, and add, modify, or delete it.

CategoryDetailed description
Tag listTag list
  • You can view the Key and Value information of the tag
  • Up to 50 tags can be added per resource
  • When entering a tag, you can search and select from the list of previously created Keys and Values
Table. GPU Node Tag Tab Items

Job History

GPU Node List page’s Job History Tab allows you to view the job history of the selected resource.

CategoryDetailed description
Task History ListResource Change History
  • Check operation details, operation time, resource type, resource name, event topic, operation result, operator information
  • Detailed Search button provides detailed search functionality
Table. GPU Node Job History Tab Detailed Information Items

Control GPU Node operation

If you need server control and management functions for the created GPU Node resources, you can perform tasks on the GPU Node List or GPU Node Details page. You can start and stop the running GPU Node resources.

Getting Started with GPU Node

You can start a stopped (Stopped) GPU Node. To start the GPU Node, follow these steps.

  1. All Services > Compute > Multi-node GPU Cluster Click the menu. 1. Go to the Service Home page of the Multi-node GPU Cluster.
  2. On the Service Home page, click the GPU Node menu. 2. Go to the GPU Node list page.
    • GPU Node List page, after selecting individual or multiple servers with the checkbox, you can Start via the More button at the top.
  3. GPU Node List page, click Resources. 3. Go to the GPU Node Details page.
    • On the GPU Node Details page, click the Start button at the top to start the server.
  4. Check the server status and complete the status change.

Stop GPU Node

You can stop a GPU node that is (Active). To stop the GPU Node, follow the steps below.

  1. All Services > Compute > Multi-node GPU Cluster Click the menu. 1. Go to the Service Home page of the Multi-node GPU Cluster.
  2. On the Service Home page, click the GPU Node menu. 2. Go to the GPU Node list page.
    • GPU Node List page, after selecting individual or multiple servers with the checkboxes, you can control them using the Stop button at the top.
  3. On the GPU Node List page, click Resources. 3. Go to the GPU Node Details page.
    • Click the 중지 button at the top of the GPU Node 상세 page to stop the server.
  4. Check the server status and complete the status change.

Terminate GPU Node

You can terminate unused GPU nodes to reduce operating costs. However, if you terminate the service, the running service may be discontinued immediately, so you should proceed with the termination only after fully considering the impact that may arise from the service interruption.

Caution
Please be aware that data cannot be recovered after terminating the service.

To cancel a GPU Node, follow these steps.

  1. All Services > Compute > Multi-node GPU Server Click the menu. 1. Go to the Service Home page of the Multi-node GPU Cluster.
  2. On the Service Home page, click the Cluster Fabric menu. 2. Go to the Cluster Fabric List page.
  3. On the Cluster Fabric List page, select the resources to cancel, and click the Cancel Service button.
    • Resources that use the same Cluster Fabric can be terminated simultaneously.
  4. When termination is complete, check on the GPU Node List page whether the resources have been terminated.
Notice

The cases where a GPU Node cannot be terminated are as follows.

  • When Block Storage(BM) is connected: Please disconnect the Block Storage(BM) connection first.
  • When File Storage is connected: Please disconnect the File Storage connection first.
  • If Lock is set: Please change the Lock setting to disabled and try again.
  • If the selection includes a server that cannot be terminated simultaneously: Please re-select only resources that can be terminated.
  • If the server you want to decommission has a different Cluster Fabric: Select only resources that use the same Cluster Fabric.
Reference
If all GPU Nodes in the Cluster Fabric are deleted, the Cluster Fabric is automatically deleted.

5.2.1 - Cluster Fabric Management

Cluster Fabric is a service that helps manage the servers (GPU Nodes) included in a GPU Cluster. Using Cluster Fabric, you can move servers between GPU clusters in the same node pool and optimize GPU performance and speed within a single GPU cluster.

Create Cluster Fabric

Cluster Fabric can be created together when a GPU Node is created, and it cannot be created or deleted separately. If all GPU Nodes in the Cluster Fabric are terminated, the Cluster Fabric is automatically deleted.
If you have not created a GPU Node, please create a GPU Node first. For more details, please refer to GPU Node 생성하기.

Check detailed information of Cluster Fabric

guide
  • Cluster Fabric can be created together when a GPU node is created, and it cannot be created or deleted independently.
  • If all GPU Nodes in the Cluster Fabric are terminated, the Cluster Fabric is automatically deleted.
  • If you have not created a GPU Node, please create a GPU Node first. * For more details, refer to GPU Node 생성하기.

On the Cluster Fabric List page and the Cluster Fabric Details page, you can view the generated Cluster Fabric list and details, and move the server.

  1. All Services > Compute > Multi-node GPU Server Click the menu. 1. Go to the Service Home page of the Multi-node GPU Cluster.

  2. On the Service Home page, click the Cluster Fabric menu. 2. Go to the Cluster Fabric List page.

    • Cluster Fabric List page allows you to view the resource list of the GPU Cluster you created.
    • Resource items other than the required columns can be added through the Settings button.
      Category
      required status
      Detailed description
      Resource IDSelectionUser-generated Cluster Fabric ID
      Cluster Fabric nameRequiredUser-created Cluster Fabric name
      Node poolSelectionA collection of nodes that can be grouped into the same Cluster Fabric
      Number of serversSelectionNumber of GPU Nodes
      Server typeSelectionServer type of GPU Node
      • The user can view the number of cores, memory capacity, and GPU type and count of the created resources
      statusSelectionStatus of the user-created Cluster Fabric
      Creation date and timeSelectionCluster Fabric creation timestamp
      Table. Cluster Fabric resource list items
  3. Cluster Fabric List page: click the resource to view detailed information. 3. Navigate to the Cluster Fabric Details page.

    • Cluster Fabric Details At the top of the page, status information and descriptions of additional features are displayed.
      CategoryDetailed description
      Cluster Fabric statusStatus of the Cluster Fabric created by the user
      • Creating: State while the cluster is being created
      • Active: State when creation is complete and it is usable
      • Editing: State while the IP is being changed
      • Deleting: State while being terminated
      • Deleted: State after termination is complete
      Add target serverA feature that allows moving a server from another cluster to the target cluster.
      Table. Cluster Fabric status information and additional functions

Detailed Information

On the Cluster Fabric List page’s Details Tab, you can view the detailed information of the selected resource and retrieve servers from another cluster.

CategoryDetailed description
serviceservice name
Resource TypeResource Type
SRNUnique resource ID in Samsung Cloud Platform
  • In Cluster Fabric, it refers to the Cluster Fabric SRN
Resource nameresource name
  • In the Cluster Fabric service, it refers to the Cluster Fabric name
Resource IDUnique resource ID in the service
ConstructorUser who created the service
Creation date and timeService creation date and time
ModifierUser who edited the service information
Modification date and timeDate and time the service information was modified
Cluster Fabric nameUser-created Cluster Fabric name
Node poolA collection of nodes that can be grouped into the same Cluster Fabric
Target serverList of GPU Nodes bound to Cluster Fabric
  • Server name, server type, IP, status
Table. Cluster Fabric detailed information tab items

Import Cluster Fabric server

Using the Add Target Server feature on the Cluster Fabric Details page, you can import servers from another cluster and add them to the selected cluster.

  1. All Services > Compute > Multi-node GPU Server Click the menu. 1. Go to the Service Home page of the Multi-node GPU Cluster.
  2. On the Service Home page, click the Cluster Fabric menu. 2. Go to the Cluster Fabric List page.
  3. Cluster Fabric List page: click the resource to view detailed information. 3. Navigate to the Cluster Fabric Details page.
  4. Click the right Add button on the target server in the Details tab.
    • The popup window for adding a target server opens. *
      • Select a cluster in Cluster Fabric.
      • GPU nodes associated with the selected cluster are listed; select the GPU node you want to retrieve.
      • The selected GPU Node’s name is displayed at the bottom.
      • Press the Confirm button to complete.
      • Pressing the Cancel button cancels the operation.
    • Check whether the added GPU Node is visible on the target server.

Terminate Cluster Fabric

If all GPU Nodes in the Cluster Fabric are terminated, the Cluster Fabric is automatically deleted. For more details, see GPU Node 해지하기.

5.2.2 - Install ServiceWatch Agent

Users can install the ServiceWatch Agent on the GPU nodes of a Multi-node GPU Cluster to collect custom metrics and logs.

Reference
Custom metric/log collection via the ServiceWatch Agent is currently available only on Samsung Cloud Platform For Enterprise. It will also be available in other offerings in the future.
Caution
Since metric collection through the ServiceWatch Agent is classified as a custom metric and incurs charges unlike the default metrics, it is recommended to remove or disable any unnecessary metric collection settings.

ServiceWatch Agent

In a Multi-node GPU Cluster, the agents that need to be installed on GPU nodes to collect ServiceWatch custom metrics and logs can be divided into two main types. It is the Prometheus Exporter and Open Telemetry Collector.

CategoryDetailed description
Prometheus ExporterProvide metrics of a specific application or service in a format that Prometheus can scrape
  • For collecting OS metrics on a GPU Node, you can use the Node Exporter for Linux servers and the Windows Exporter for Windows servers, depending on the OS type.
Open Telemetry CollectorActs as a centralized collector that gathers telemetry data such as metrics and logs from distributed systems, processes (filtering, sampling, etc.) it, and exports it to various backends (e.g., Prometheus, Jaeger, Elasticsearch, etc.)
  • Exports data via the ServiceWatch Gateway so that ServiceWatch can collect metric and log data.
Table. Explanation of Prometheus Exporter and Open Telemetry Collector
information

If you have configured a Kubernetes Engine on a GPU node, please view the GPU metrics using the metrics provided by the Kubernetes Engine.

  • If you install the DCGM Exporter on a GPU node configured with Kubernetes Engine, it may not operate correctly.
Reference
The ServiceWatch Agent guide for collecting GPU metrics on a GPU Node can be used the same way as on a GPU Server. For more details, see GPU Server > ServiceWatch Agent.

Pre-configuration for Using ServiceWatch Agent

To use the ServiceWatch Agent, please refer to ServiceWatch Agent를 위한 사전 환경 설정 and prepare the prerequisite configuration.

5.2.3 - Multi-node GPU Cluster Service Scope and Inspection Guide

Multi-node GPU Cluster Service Scope

If an IaaS hardware-level issue occurs with the Multi-node GPU Cluster service, you can receive technical support through Contact Us in the Support Center. However, the risks associated with changes such as OS kernel updates or application installations are the user’s responsibility, so technical support is limited; please be mindful when performing tasks such as system updates.

IaaS hardware level issue

  • HW fault event messages generated within the server by the IPMI hardware monitoring console.
  • GPU HW operation error observed in the nvdia-smi command
  • HW error messages that occur during inspection of InfiniBand HCA cards or InfiniBand Switches
Caution
Since the Multi-node GPU Cluster is a service sensitive to software version compatibility of Ubuntu OS / NVDIA / Infiniband, official technical support is unavailable after changes such as a user’s OS kernel update or application installation.

IaaS HW Inspection Guide

After applying for the Multi-node GPU Cluster service, it is recommended to check the IaaS HW level according to the inspection guide.

Intel E810 driver update

Check the version of the Intel E810 driver and, if necessary, refer to the following procedure to perform an update.

Caution
Proceed with the update only if the Multi-node GPU Cluster Node has an Intel E810 Device and is simultaneously using a standard image version 535.86.10 or lower.
Reference

You can use the lspci command to verify whether an E810 NIC device is present. If the E810 NIC is valid, the PCIe device is identified as follows using the E810-C information (if there is no E810 NIC device, this operation is not performed).

<div class="code-block-buttons">
  
  <button class="code-block-download">
    <div class="code-block-download-icon">
      <svg width="14" height="16" viewBox="0 0 14 16" fill="none" xmlns="http://www.w3.org/2000/svg"><path d="M13.999 14.4353v-2.998C13.999 11.0232 13.6602 10.6853 13.2461 10.6853S12.5 11.0232 12.5 11.4373v2.2519H1.5V11.4373C1.5 11.0232 1.16211 10.6853.748047 10.6853.333984 10.6853 976563e-9 11.0232 976562e-9 11.4373v2.998C976562e-9 14.8494.333984 15.1892.748047 15.1892H13.2461c.414099999999999.0.7529-.3398.7529-.7539z" fill="#5135ff"/><path d="M1.41169 6.21654c.25876-.32345.73073-.37589 1.05417-.11713l3.78463 3.0277V1.56104c0-.41422.33578-.750005.75-.750005.41421.0.75.335785.75.750005V9.12208L11.5288 6.09941C11.8523 5.84065 12.3242 5.89309 12.583 6.21654 12.8418 6.53999 12.7893 7.01196 12.4659 7.27071L6.99734 11.6455 1.52882 7.27071c-.32345-.25875-.37589-.73072-.11713-1.05417z" fill="#5135ff"/></svg></div>
    Sample Code Download
  </button>
  
  <button class="code-block-copy">
    <div class="code-block-copy-icon">
      <svg width="14" height="14" viewBox="0 0 14 14" fill="none" xmlns="http://www.w3.org/2000/svg"><path fill-rule="evenodd" clip-rule="evenodd" d="M10 4.00012V1c0-.552285-.44772-1-1-1H1C.447715.0.0.447715.0 1V9c0 .55228.447715 1 1 1H3.99988v3.0001C3.99988 13.5515 4.44849 14 4.99988 14H10.0002l3.9997-3.9995V5c0-.55139-.448600000000001-.99988-1-.99988H10zM1.4 1.4V8.6H3.99988V5c0-.55139.44861-.99988 1-.99988H8.6V1.4H1.4zM5.3999 12.6H9.08295V9.78301C9.08295 9.39641 9.39635 9.08301 9.78295 9.08301H12.5999V5.40015h-7.2V12.6zm6.1375-2.117-1.0545 1.0544V10.483h1.0545z" fill="currentcolor"/></svg></div>
    Copy Code
  </button>
</div>
Color mode
lspci | grep E810
0000:6a:00.0 Ethernet controller: Intel Corporation Ethernet Controller E810-C for QSFP (rev 02) 0000:6a:00.1 Ethernet controller: Intel Corporation Ethernet Controller E810-C for QSFP (rev 02)
lspci | grep E810
0000:6a:00.0 Ethernet controller: Intel Corporation Ethernet Controller E810-C for QSFP (rev 02) 0000:6a:00.1 Ethernet controller: Intel Corporation Ethernet Controller E810-C for QSFP (rev 02)
</div>
Example of checking for the presence of an E810 NIC device

To update the driver, follow the steps below.

  1. Move the base driver tar file to the desired directory.

    Color mode
    /usr/local/src
    /usr/local/src
    Directory navigation example

  2. Untar / unzip the Archiver file.

    • x.x.x is the version number of the driver tar file.
      Color mode
      tar zxf ice-x.x.x.tar.gz
      tar zxf ice-x.x.x.tar.gz
      untar example
  3. Change the driver to the src directory.

    • x.x.x is the version number of the driver tar file.
      Color mode
      cd ice-x.x.x/src/
      cd ice-x.x.x/src/
      Example of changing to the src directory
  4. Compile the driver module.

    Color mode
    make install
    make install
    Driver module compilation example

  5. After the update is complete, check the version.

    Color mode
    lsmod | grep ice
    modinfo ice | grep version
    lsmod | grep ice
    modinfo ice | grep version
    Version check example

Check NVIDIA driver

To check the NVIDIA driver (nvidia-smi topo, IB nv_peer_mem status) and inspect the IaaS hardware level, follow these steps.

Reference
The example of applying MIG is described using an A100 GPU node as the reference.
  1. Check the GPU driver status.

    Color mode
    ~$ nvidia-smi
    ~$ nvidia-smi
    Example code for checking GPU driver status
    Color mode
    Thu Jan 29 14:48:31 2026
    +---------------------------------------------------------------------------------------+
    | NVIDIA-SMI 535.183.06             Driver Version: 535.183.06   CUDA Version: 12.2 |
    |-----------------------------------------+----------------------+----------------------+
    | GPU  Name                 Persistence-M | Bus-Id        Disp.A | Volatile Uncorr. ECC |
    | Fan  Temp   Perf          Pwr:Usage/Cap | Memory-Usage | GPU-Util  Compute M. |
    |  |  | MIG M. |
    | =========================================+======================+====================== |
    | 0  NVIDIA A100-SXM4-80GB          On | 00000000:00:05.0 Off | On |
    | N/A   36C    P0              52W / 400W | 0MiB / 81920MiB | N/A      Default |
    |  |  | Enabled |
    +-----------------------------------------+----------------------+----------------------+
    | 1  NVIDIA A100-SXM4-80GB          On | 00000000:00:06.0 Off | 0 |
    | N/A   36C    P0              61W / 400W | 0MiB / 81920MiB | 0%      Default |
    |  |  | Disabled |
    +-----------------------------------------+----------------------+----------------------+
    | 2  NVIDIA A100-SXM4-80GB          On | 00000000:00:07.0 Off | 0 |
    | N/A   36C    P0              64W / 400W | 0MiB / 81920MiB | 0%      Default |
    |  |  | Disabled |
    +-----------------------------------------+----------------------+----------------------+
    | 3  NVIDIA A100-SXM4-80GB          On | 00000000:00:08.0 Off | 0 |
    | N/A   40C    P0              64W / 400W | 0MiB / 81920MiB | 0%      Default |
    |  |  | Disabled |
    +-----------------------------------------+----------------------+----------------------+
    | 4  NVIDIA A100-SXM4-80GB          On | 00000000:00:09.0 Off | 0 |
    | N/A   36C    P0              63W / 400W | 0MiB / 81920MiB | 0%      Default |
    |  |  | Disabled |
    +-----------------------------------------+----------------------+----------------------+
    | 5  NVIDIA A100-SXM4-80GB          On | 00000000:00:0A.0 Off | 0 |
    | N/A   40C    P0              64W / 400W | 0MiB / 81920MiB | 0%      Default |
    |  |  | Disabled |
    +-----------------------------------------+----------------------+----------------------+
    | 6  NVIDIA A100-SXM4-80GB          On | 00000000:00:0B.0 Off | 0 |
    | N/A   39C    P0              65W / 400W | 0MiB / 81920MiB | 0%      Default |
    |  |  | Disabled |
    +-----------------------------------------+----------------------+----------------------+
    | 7  NVIDIA A100-SXM4-80GB          On | 00000000:00:0C.0 Off | 0 |
    | N/A   39C    P0              60W / 400W | 0MiB / 81920MiB | 0%      Default |
    |  |  | Disabled |
    +-----------------------------------------+----------------------+----------------------+
    
    +---------------------------------------------------------------------------------------+
    | MIG devices: |
    +------------------+--------------------------------+-----------+-----------------------+
    | GPU  GI  CI  MIG | Memory-Usage | Vol | Shared |
    | ID  ID  Dev | BAR1-Usage | SM     Unc | CE ENC DEC OFA JPG |
    |  |  | ECC |  |
    | ==================+================================+===========+======================= |
    | No MIG devices found |
    +---------------------------------------------------------------------------------------+
    
    +---------------------------------------------------------------------------------------+
    | Processes: |
    | GPU   GI   CI        PID   Type   Process name                            GPU Memory |
    | ID   ID                                                             Usage |
    | ======================================================================================= |
    | No running processes found |
    +---------------------------------------------------------------------------------------+
    Thu Jan 29 14:48:31 2026
    +---------------------------------------------------------------------------------------+
    | NVIDIA-SMI 535.183.06             Driver Version: 535.183.06   CUDA Version: 12.2 |
    |-----------------------------------------+----------------------+----------------------+
    | GPU  Name                 Persistence-M | Bus-Id        Disp.A | Volatile Uncorr. ECC |
    | Fan  Temp   Perf          Pwr:Usage/Cap | Memory-Usage | GPU-Util  Compute M. |
    |  |  | MIG M. |
    | =========================================+======================+====================== |
    | 0  NVIDIA A100-SXM4-80GB          On | 00000000:00:05.0 Off | On |
    | N/A   36C    P0              52W / 400W | 0MiB / 81920MiB | N/A      Default |
    |  |  | Enabled |
    +-----------------------------------------+----------------------+----------------------+
    | 1  NVIDIA A100-SXM4-80GB          On | 00000000:00:06.0 Off | 0 |
    | N/A   36C    P0              61W / 400W | 0MiB / 81920MiB | 0%      Default |
    |  |  | Disabled |
    +-----------------------------------------+----------------------+----------------------+
    | 2  NVIDIA A100-SXM4-80GB          On | 00000000:00:07.0 Off | 0 |
    | N/A   36C    P0              64W / 400W | 0MiB / 81920MiB | 0%      Default |
    |  |  | Disabled |
    +-----------------------------------------+----------------------+----------------------+
    | 3  NVIDIA A100-SXM4-80GB          On | 00000000:00:08.0 Off | 0 |
    | N/A   40C    P0              64W / 400W | 0MiB / 81920MiB | 0%      Default |
    |  |  | Disabled |
    +-----------------------------------------+----------------------+----------------------+
    | 4  NVIDIA A100-SXM4-80GB          On | 00000000:00:09.0 Off | 0 |
    | N/A   36C    P0              63W / 400W | 0MiB / 81920MiB | 0%      Default |
    |  |  | Disabled |
    +-----------------------------------------+----------------------+----------------------+
    | 5  NVIDIA A100-SXM4-80GB          On | 00000000:00:0A.0 Off | 0 |
    | N/A   40C    P0              64W / 400W | 0MiB / 81920MiB | 0%      Default |
    |  |  | Disabled |
    +-----------------------------------------+----------------------+----------------------+
    | 6  NVIDIA A100-SXM4-80GB          On | 00000000:00:0B.0 Off | 0 |
    | N/A   39C    P0              65W / 400W | 0MiB / 81920MiB | 0%      Default |
    |  |  | Disabled |
    +-----------------------------------------+----------------------+----------------------+
    | 7  NVIDIA A100-SXM4-80GB          On | 00000000:00:0C.0 Off | 0 |
    | N/A   39C    P0              60W / 400W | 0MiB / 81920MiB | 0%      Default |
    |  |  | Disabled |
    +-----------------------------------------+----------------------+----------------------+
    
    +---------------------------------------------------------------------------------------+
    | MIG devices: |
    +------------------+--------------------------------+-----------+-----------------------+
    | GPU  GI  CI  MIG | Memory-Usage | Vol | Shared |
    | ID  ID  Dev | BAR1-Usage | SM     Unc | CE ENC DEC OFA JPG |
    |  |  | ECC |  |
    | ==================+================================+===========+======================= |
    | No MIG devices found |
    +---------------------------------------------------------------------------------------+
    
    +---------------------------------------------------------------------------------------+
    | Processes: |
    | GPU   GI   CI        PID   Type   Process name                            GPU Memory |
    | ID   ID                                                             Usage |
    | ======================================================================================= |
    | No running processes found |
    +---------------------------------------------------------------------------------------+
    GPU driver status example

  2. Check the NVSwitch and NVLink hardware status.

    • Check NVSwitch status

      Color mode
      ~$ nvidia-smi nvlink --status
      ~$ nvidia-smi nvlink --status
      NVSwitch status check example
      Color mode
      GPU 1: NVIDIA A100-SXM4-80GB (UUID: GPU-64a2f685-bb12-c4af-105c-0726ece9c8d7)
               Link 0: 25 GB/s
               Link 1: 25 GB/s
               Link 2: 25 GB/s
               Link 3: 25 GB/s
               Link 4: 25 GB/s
               Link 5: 25 GB/s
               Link 6: 25 GB/s
               Link 7: 25 GB/s
               Link 8: 25 GB/s
               Link 9: 25 GB/s
               Link 10: 25 GB/s
               Link 11: 25 GB/s
      GPU 2: NVIDIA A100-SXM4-80GB (UUID: GPU-2269851b-71cd-f6c7-50c5-ba1525cf3ce8)
               Link 0: 25 GB/s
               Link 1: 25 GB/s
               Link 2: 25 GB/s
               Link 3: 25 GB/s
               Link 4: 25 GB/s
               Link 5: 25 GB/s
               Link 6: 25 GB/s
               Link 7: 25 GB/s
               Link 8: 25 GB/s
               Link 9: 25 GB/s
               Link 10: 25 GB/s
               Link 11: 25 GB/s
      GPU 3: NVIDIA A100-SXM4-80GB (UUID: GPU-4c397bbf-95fc-5c29-918a-a429cbe45a7a)
               Link 0: 25 GB/s
               Link 1: 25 GB/s
               Link 2: 25 GB/s
               Link 3: 25 GB/s
               Link 4: 25 GB/s
               Link 5: 25 GB/s
               Link 6: 25 GB/s
               Link 7: 25 GB/s
               Link 8: 25 GB/s
               Link 9: 25 GB/s
               Link 10: 25 GB/s
               Link 11: 25 GB/s
      GPU 4: NVIDIA A100-SXM4-80GB (UUID: GPU-0e350204-9fb6-2cbe-538e-8f7849658eb8)
               Link 0: 25 GB/s
               Link 1: 25 GB/s
               Link 2: 25 GB/s
               Link 3: 25 GB/s
               Link 4: 25 GB/s
               Link 5: 25 GB/s
               Link 6: 25 GB/s
               Link 7: 25 GB/s
               Link 8: 25 GB/s
               Link 9: 25 GB/s
               Link 10: 25 GB/s
               Link 11: 25 GB/s
      GPU 5: NVIDIA A100-SXM4-80GB (UUID: GPU-45f0c453-4760-edd4-3af9-25c5ea7473a5)
               Link 0: 25 GB/s
               Link 1: 25 GB/s
               Link 2: 25 GB/s
               Link 3: 25 GB/s
               Link 4: 25 GB/s
               Link 5: 25 GB/s
               Link 6: 25 GB/s
               Link 7: 25 GB/s
               Link 8: 25 GB/s
               Link 9: 25 GB/s
               Link 10: 25 GB/s
               Link 11: 25 GB/s
      GPU 6: NVIDIA A100-SXM4-80GB (UUID: GPU-38409794-bb34-430e-3c50-90b42cb2bb72)
               Link 0: 25 GB/s
               Link 1: 25 GB/s
               Link 2: 25 GB/s
               Link 3: 25 GB/s
               Link 4: 25 GB/s
               Link 5: 25 GB/s
               Link 6: 25 GB/s
               Link 7: 25 GB/s
               Link 8: 25 GB/s
               Link 9: 25 GB/s
               Link 10: 25 GB/s
               Link 11: 25 GB/s
      GPU 7: NVIDIA A100-SXM4-80GB (UUID: GPU-3fb478aa-801b-eb64-55c2-0ffc3f2ce404)
               Link 0: 25 GB/s
               Link 1: 25 GB/s
               Link 2: 25 GB/s
               Link 3: 25 GB/s
               Link 4: 25 GB/s
               Link 5: 25 GB/s
               Link 6: 25 GB/s
               Link 7: 25 GB/s
               Link 8: 25 GB/s
               Link 9: 25 GB/s
               Link 10: 25 GB/s
               Link 11: 25 GB/s
      GPU 1: NVIDIA A100-SXM4-80GB (UUID: GPU-64a2f685-bb12-c4af-105c-0726ece9c8d7)
               Link 0: 25 GB/s
               Link 1: 25 GB/s
               Link 2: 25 GB/s
               Link 3: 25 GB/s
               Link 4: 25 GB/s
               Link 5: 25 GB/s
               Link 6: 25 GB/s
               Link 7: 25 GB/s
               Link 8: 25 GB/s
               Link 9: 25 GB/s
               Link 10: 25 GB/s
               Link 11: 25 GB/s
      GPU 2: NVIDIA A100-SXM4-80GB (UUID: GPU-2269851b-71cd-f6c7-50c5-ba1525cf3ce8)
               Link 0: 25 GB/s
               Link 1: 25 GB/s
               Link 2: 25 GB/s
               Link 3: 25 GB/s
               Link 4: 25 GB/s
               Link 5: 25 GB/s
               Link 6: 25 GB/s
               Link 7: 25 GB/s
               Link 8: 25 GB/s
               Link 9: 25 GB/s
               Link 10: 25 GB/s
               Link 11: 25 GB/s
      GPU 3: NVIDIA A100-SXM4-80GB (UUID: GPU-4c397bbf-95fc-5c29-918a-a429cbe45a7a)
               Link 0: 25 GB/s
               Link 1: 25 GB/s
               Link 2: 25 GB/s
               Link 3: 25 GB/s
               Link 4: 25 GB/s
               Link 5: 25 GB/s
               Link 6: 25 GB/s
               Link 7: 25 GB/s
               Link 8: 25 GB/s
               Link 9: 25 GB/s
               Link 10: 25 GB/s
               Link 11: 25 GB/s
      GPU 4: NVIDIA A100-SXM4-80GB (UUID: GPU-0e350204-9fb6-2cbe-538e-8f7849658eb8)
               Link 0: 25 GB/s
               Link 1: 25 GB/s
               Link 2: 25 GB/s
               Link 3: 25 GB/s
               Link 4: 25 GB/s
               Link 5: 25 GB/s
               Link 6: 25 GB/s
               Link 7: 25 GB/s
               Link 8: 25 GB/s
               Link 9: 25 GB/s
               Link 10: 25 GB/s
               Link 11: 25 GB/s
      GPU 5: NVIDIA A100-SXM4-80GB (UUID: GPU-45f0c453-4760-edd4-3af9-25c5ea7473a5)
               Link 0: 25 GB/s
               Link 1: 25 GB/s
               Link 2: 25 GB/s
               Link 3: 25 GB/s
               Link 4: 25 GB/s
               Link 5: 25 GB/s
               Link 6: 25 GB/s
               Link 7: 25 GB/s
               Link 8: 25 GB/s
               Link 9: 25 GB/s
               Link 10: 25 GB/s
               Link 11: 25 GB/s
      GPU 6: NVIDIA A100-SXM4-80GB (UUID: GPU-38409794-bb34-430e-3c50-90b42cb2bb72)
               Link 0: 25 GB/s
               Link 1: 25 GB/s
               Link 2: 25 GB/s
               Link 3: 25 GB/s
               Link 4: 25 GB/s
               Link 5: 25 GB/s
               Link 6: 25 GB/s
               Link 7: 25 GB/s
               Link 8: 25 GB/s
               Link 9: 25 GB/s
               Link 10: 25 GB/s
               Link 11: 25 GB/s
      GPU 7: NVIDIA A100-SXM4-80GB (UUID: GPU-3fb478aa-801b-eb64-55c2-0ffc3f2ce404)
               Link 0: 25 GB/s
               Link 1: 25 GB/s
               Link 2: 25 GB/s
               Link 3: 25 GB/s
               Link 4: 25 GB/s
               Link 5: 25 GB/s
               Link 6: 25 GB/s
               Link 7: 25 GB/s
               Link 8: 25 GB/s
               Link 9: 25 GB/s
               Link 10: 25 GB/s
               Link 11: 25 GB/s
      NVSwitch status example

    • Check NVLinks hardware status

      Color mode
      ~$ nvidia-smi topo -m
      ~$ nvidia-smi topo -m
      Example code for checking NVLink hardware status
      Color mode
              GPU0    GPU1    GPU2    GPU3    GPU4    GPU5    GPU6    GPU7    CPU Affinity    NUMA Affinity   GPU NUMA ID
      GPU0     X      NV12    NV12    NV12    NV12    NV12    NV12    NV12    0-127   0-7             N/A
      GPU1    NV12	 X      NV12    NV12    NV12    NV12    NV12    NV12    0-127   0-7             N/A
      GPU2    NV12    NV12     X      NV12    NV12    NV12    NV12    NV12    0-127   0-7             N/A
      GPU3    NV12    NV12    NV12     X      NV12    NV12    NV12    NV12    0-127   0-7             N/A
      GPU4    NV12    NV12    NV12    NV12     X      NV12    NV12    NV12    0-127   0-7             N/A
      GPU5    NV12    NV12    NV12    NV12    NV12     X      NV12    NV12    0-127   0-7             N/A
      GPU6    NV12    NV12    NV12    NV12    NV12    NV12     X      NV12    0-127   0-7             N/A
      GPU7    NV12    NV12    NV12    NV12    NV12    NV12    NV12     X      0-127   0-7             N/A
      
      Legend:
      
        X    = Self
        SYS  = Connection traversing PCIe as well as the SMP interconnect between NUMA nodes (e.g., QPI/UPI)
        NODE = Connection traversing PCIe as well as the interconnect between PCIe Host Bridges within a NUMA node
        PHB  = Connection traversing PCIe as well as a PCIe Host Bridge (typically the CPU)
        PXB  = Connection traversing multiple PCIe bridges (without traversing the PCIe Host Bridge)
        PIX  = Connection traversing at most a single PCIe bridge
        NV#  = Connection traversing a bonded set of # NVLinks
              GPU0    GPU1    GPU2    GPU3    GPU4    GPU5    GPU6    GPU7    CPU Affinity    NUMA Affinity   GPU NUMA ID
      GPU0     X      NV12    NV12    NV12    NV12    NV12    NV12    NV12    0-127   0-7             N/A
      GPU1    NV12	 X      NV12    NV12    NV12    NV12    NV12    NV12    0-127   0-7             N/A
      GPU2    NV12    NV12     X      NV12    NV12    NV12    NV12    NV12    0-127   0-7             N/A
      GPU3    NV12    NV12    NV12     X      NV12    NV12    NV12    NV12    0-127   0-7             N/A
      GPU4    NV12    NV12    NV12    NV12     X      NV12    NV12    NV12    0-127   0-7             N/A
      GPU5    NV12    NV12    NV12    NV12    NV12     X      NV12    NV12    0-127   0-7             N/A
      GPU6    NV12    NV12    NV12    NV12    NV12    NV12     X      NV12    0-127   0-7             N/A
      GPU7    NV12    NV12    NV12    NV12    NV12    NV12    NV12     X      0-127   0-7             N/A
      
      Legend:
      
        X    = Self
        SYS  = Connection traversing PCIe as well as the SMP interconnect between NUMA nodes (e.g., QPI/UPI)
        NODE = Connection traversing PCIe as well as the interconnect between PCIe Host Bridges within a NUMA node
        PHB  = Connection traversing PCIe as well as a PCIe Host Bridge (typically the CPU)
        PXB  = Connection traversing multiple PCIe bridges (without traversing the PCIe Host Bridge)
        PIX  = Connection traversing at most a single PCIe bridge
        NV#  = Connection traversing a bonded set of # NVLinks
      NVLink HW status check code example

  3. Check the InfiniBand (IB) HCA card hardware status and link.

    Color mode
    user@bm-dev-001:~$ ibdev2netdev -v
    user@bm-dev-001:~$ ibdev2netdev -v
    HW status check command example
    Color mode
    cat: /sys/class/infiniband/mlx5_0/device/vpd: Permission denied
    0000:45:00.0 mlx5_0 (MT4123 -            )                 fw 20.29.1016 port 1 (ACTIVE) ==> ibs18 (Down)
    cat: /sys/class/infiniband/mlx5_1/device/vpd: Permission denied
    0000:0e:00.0 mlx5_1 (MT4123 -            )                 fw 20.29.1016 port 1 (ACTIVE) ==> ibs17 (Down)
    cat: /sys/class/infiniband/mlx5_2/device/vpd: Permission denied
    0000:c5:00.0 mlx5_2 (MT4123 -            )                 fw 20.29.1016 port 1 (ACTIVE) ==> ibs20 (Down)
    cat: /sys/class/infiniband/mlx5_3/device/vpd: Permission denied
    0000:85:00.0 mlx5_3 (MT4123 -            )                 fw 20.29.1016 port 1 (ACTIVE) ==> ibs19 (Down)
    user@bm-dev-001:~$
    cat: /sys/class/infiniband/mlx5_0/device/vpd: Permission denied
    0000:45:00.0 mlx5_0 (MT4123 -            )                 fw 20.29.1016 port 1 (ACTIVE) ==> ibs18 (Down)
    cat: /sys/class/infiniband/mlx5_1/device/vpd: Permission denied
    0000:0e:00.0 mlx5_1 (MT4123 -            )                 fw 20.29.1016 port 1 (ACTIVE) ==> ibs17 (Down)
    cat: /sys/class/infiniband/mlx5_2/device/vpd: Permission denied
    0000:c5:00.0 mlx5_2 (MT4123 -            )                 fw 20.29.1016 port 1 (ACTIVE) ==> ibs20 (Down)
    cat: /sys/class/infiniband/mlx5_3/device/vpd: Permission denied
    0000:85:00.0 mlx5_3 (MT4123 -            )                 fw 20.29.1016 port 1 (ACTIVE) ==> ibs19 (Down)
    user@bm-dev-001:~$
    Example of HW status check result
    Color mode
    root@bm-dev-001:~# ibstat
    root@bm-dev-001:~# ibstat
    Example of link verification command
    Color mode
    CA 'mlx5_0'
            CA type: MT4123
            Number of ports: 1
            Firmware version: 20.29.1016
            Hardware version: 0
            Node GUID: 0x88e9a4ffff5060ac
            System image GUID: 0x88e9a4ffff5060ac
            Port 1:
                    State: Active
                    Physical state: LinkUp
                    Rate: 200
                    Base lid: 8
                    LMC: 0
                    SM lid: 1
                    Capability mask: 0x2651e848
                    Port GUID: 0x88e9a4ffff5060ac
                    Link layer: InfiniBand
    CA 'mlx5_1'
            CA type: MT4123
            Number of ports: 1
            Firmware version: 20.29.1016
            Hardware version: 0
            Node GUID: 0x88e9a4ffff504080
            System image GUID: 0x88e9a4ffff504080
            Port 1:
                    State: Active
                    Physical state: LinkUp
                    Rate: 200
                    Base lid: 5
                    LMC: 0
                    SM lid: 1
                    Capability mask: 0x2651e848
                    Port GUID: 0x88e9a4ffff504080
                    Link layer: InfiniBand
    CA 'mlx5_2'
            CA type: MT4123
            Number of ports: 1
            Firmware version: 20.29.1016
            Hardware version: 0
            Node GUID: 0x88e9a4ffff505038
            System image GUID: 0x88e9a4ffff505038
            Port 1:
                    State: Active
                    Physical state: LinkUp
                    Rate: 200
                    Base lid: 2
                    LMC: 0
                    SM lid: 1
                    Capability mask: 0x2651e848
                    Port GUID: 0x88e9a4ffff505038
                    Link layer: InfiniBand
    CA 'mlx5_3'
            CA type: MT4123
            Number of ports: 1
            Firmware version: 20.29.1016
            Hardware version: 0
            Node GUID: 0x88e9a4ffff504094
            System image GUID: 0x88e9a4ffff504094
            Port 1:
                    State: Active
                    Physical state: LinkUp
                    Rate: 200
                    Base lid: 7
                    LMC: 0
                    SM lid: 1
                    Capability mask: 0x2651e848
                    Port GUID: 0x88e9a4ffff504094
                    Link layer: InfiniBand
    CA 'mlx5_0'
            CA type: MT4123
            Number of ports: 1
            Firmware version: 20.29.1016
            Hardware version: 0
            Node GUID: 0x88e9a4ffff5060ac
            System image GUID: 0x88e9a4ffff5060ac
            Port 1:
                    State: Active
                    Physical state: LinkUp
                    Rate: 200
                    Base lid: 8
                    LMC: 0
                    SM lid: 1
                    Capability mask: 0x2651e848
                    Port GUID: 0x88e9a4ffff5060ac
                    Link layer: InfiniBand
    CA 'mlx5_1'
            CA type: MT4123
            Number of ports: 1
            Firmware version: 20.29.1016
            Hardware version: 0
            Node GUID: 0x88e9a4ffff504080
            System image GUID: 0x88e9a4ffff504080
            Port 1:
                    State: Active
                    Physical state: LinkUp
                    Rate: 200
                    Base lid: 5
                    LMC: 0
                    SM lid: 1
                    Capability mask: 0x2651e848
                    Port GUID: 0x88e9a4ffff504080
                    Link layer: InfiniBand
    CA 'mlx5_2'
            CA type: MT4123
            Number of ports: 1
            Firmware version: 20.29.1016
            Hardware version: 0
            Node GUID: 0x88e9a4ffff505038
            System image GUID: 0x88e9a4ffff505038
            Port 1:
                    State: Active
                    Physical state: LinkUp
                    Rate: 200
                    Base lid: 2
                    LMC: 0
                    SM lid: 1
                    Capability mask: 0x2651e848
                    Port GUID: 0x88e9a4ffff505038
                    Link layer: InfiniBand
    CA 'mlx5_3'
            CA type: MT4123
            Number of ports: 1
            Firmware version: 20.29.1016
            Hardware version: 0
            Node GUID: 0x88e9a4ffff504094
            System image GUID: 0x88e9a4ffff504094
            Port 1:
                    State: Active
                    Physical state: LinkUp
                    Rate: 200
                    Base lid: 7
                    LMC: 0
                    SM lid: 1
                    Capability mask: 0x2651e848
                    Port GUID: 0x88e9a4ffff504094
                    Link layer: InfiniBand
    Link verification result example

Check IB bandwidth communication

Check the IB bandwidth communication status (ib_send_bw) and follow the steps below to inspect the IaaS hardware level.

  1. Check the name of the IB HCA interface.

    • In the following example, IB ports: mlx5_0, mlx5_4, mlx_5_5, mlx5_8
      Color mode
      ~$ ibdev2netdev -v
      ~$ ibdev2netdev -v
      Example of checking IB HCA interface name
      Color mode
      0000:1a:00.0 mlx5_0 (MT4129 - 1028SN     ) Nvidia ConnectX-7 Single Port Infiniband NDR OSFP Adapter fw 28.38.1002 port 1 (ACTIVE) ==> ibp26s0 (Down)
      0000:1b:00.0 mlx5_1 (MT4123 - 1028SN     ) Mellanox ConnectX-6 Single Port VPI HDR QSFP Adapter fw 20.38.1002 port 1 (ACTIVE) ==> bond-nas (Up)
      0000:3c:00.0 mlx5_2 (MT4129 - 1028SN     ) Nvidia ConnectX-7 Single Port Infiniband NDR OSFP Adapter fw 28.38.1002 port 1 (DOWN  ) ==> ibp60s0 (Down)
      0000:4d:00.0 mlx5_3 (MT4129 - 1028SN     ) Nvidia ConnectX-7 Single Port Infiniband NDR OSFP Adapter fw 28.38.1002 port 1 (DOWN  ) ==> ibp77s0 (Down)
      0000:5e:00.0 mlx5_4 (MT4129 - 1028SN     ) Nvidia ConnectX-7 Single Port Infiniband NDR OSFP Adapter fw 28.38.1002 port 1 (ACTIVE) ==> ibp94s0 (Down)
      0000:9c:00.0 mlx5_5 (MT4129 - 1028SN     ) Nvidia ConnectX-7 Single Port Infiniband NDR OSFP Adapter fw 28.38.1002 port 1 (ACTIVE) ==> ibp156s0 (Down)
      0000:9d:00.0 mlx5_6 (MT4123 - 1028SN     ) Mellanox ConnectX-6 Single Port VPI HDR QSFP Adapter fw 20.38.1002 port 1 (ACTIVE) ==> bond-nas (Up)
      0000:bc:00.0 mlx5_7 (MT4129 - 1028SN     ) Nvidia ConnectX-7 Single Port Infiniband NDR OSFP Adapter fw 28.38.1002 port 1 (DOWN  ) ==> ibp188s0 (Down)
      0000:cc:00.0 mlx5_8 (MT4129 - 1028SN     ) Nvidia ConnectX-7 Single Port Infiniband NDR OSFP Adapter fw 28.38.1002 port 1 (ACTIVE) ==> ibp204s0 (Down)
      0000:dc:00.0 mlx5_9 (MT4129 - 1028SN     ) Nvidia ConnectX-7 Single Port Infiniband NDR OSFP Adapter fw 28.38.1002 port 1 (DOWN  ) ==> ibp220s0 (Down)
      0000:1a:00.0 mlx5_0 (MT4129 - 1028SN     ) Nvidia ConnectX-7 Single Port Infiniband NDR OSFP Adapter fw 28.38.1002 port 1 (ACTIVE) ==> ibp26s0 (Down)
      0000:1b:00.0 mlx5_1 (MT4123 - 1028SN     ) Mellanox ConnectX-6 Single Port VPI HDR QSFP Adapter fw 20.38.1002 port 1 (ACTIVE) ==> bond-nas (Up)
      0000:3c:00.0 mlx5_2 (MT4129 - 1028SN     ) Nvidia ConnectX-7 Single Port Infiniband NDR OSFP Adapter fw 28.38.1002 port 1 (DOWN  ) ==> ibp60s0 (Down)
      0000:4d:00.0 mlx5_3 (MT4129 - 1028SN     ) Nvidia ConnectX-7 Single Port Infiniband NDR OSFP Adapter fw 28.38.1002 port 1 (DOWN  ) ==> ibp77s0 (Down)
      0000:5e:00.0 mlx5_4 (MT4129 - 1028SN     ) Nvidia ConnectX-7 Single Port Infiniband NDR OSFP Adapter fw 28.38.1002 port 1 (ACTIVE) ==> ibp94s0 (Down)
      0000:9c:00.0 mlx5_5 (MT4129 - 1028SN     ) Nvidia ConnectX-7 Single Port Infiniband NDR OSFP Adapter fw 28.38.1002 port 1 (ACTIVE) ==> ibp156s0 (Down)
      0000:9d:00.0 mlx5_6 (MT4123 - 1028SN     ) Mellanox ConnectX-6 Single Port VPI HDR QSFP Adapter fw 20.38.1002 port 1 (ACTIVE) ==> bond-nas (Up)
      0000:bc:00.0 mlx5_7 (MT4129 - 1028SN     ) Nvidia ConnectX-7 Single Port Infiniband NDR OSFP Adapter fw 28.38.1002 port 1 (DOWN  ) ==> ibp188s0 (Down)
      0000:cc:00.0 mlx5_8 (MT4129 - 1028SN     ) Nvidia ConnectX-7 Single Port Infiniband NDR OSFP Adapter fw 28.38.1002 port 1 (ACTIVE) ==> ibp204s0 (Down)
      0000:dc:00.0 mlx5_9 (MT4129 - 1028SN     ) Nvidia ConnectX-7 Single Port Infiniband NDR OSFP Adapter fw 28.38.1002 port 1 (DOWN  ) ==> ibp220s0 (Down)
      Example of IB HCA interface name verification result
  2. Use the SERVER Side command to check the communication status.

    Color mode
    ~$ ib_send_bw -d mlx5_0 -i 1 –F
    ~$ ib_send_bw -d mlx5_0 -i 1 –F
    SERVER Side command example
    Color mode
    ************************************
    * Waiting for client to connect... * *
    ************************************
    ---------------------------------------------------------------------------------------
                        Send BW Test
     Dual-port       : OFF		Device         : mlx5_0
     Number of qps   : 1		Transport type : IB
     Connection type : RC		Using SRQ      : OFF
     PCIe relax order: ON
     ibv_wr* API     : ON
     TX depth        : 128
     CQ Moderation   : 100
     Mtu             : 4096[B]
     Link type       : IB
     Max inline data : 0[B]
     rdma_cm QPs	 : OFF
     Data ex.  method : Ethernet
    ---------------------------------------------------------------------------------------
     local address: LID 0x0f QPN 0x6d95 PSN 0xb974a
     remote address: LID 0x01 QPN 0x6dd2 PSN 0xc8a18c
    ---------------------------------------------------------------------------------------
     #bytes     #iterations    BW peak[MB/sec]    BW average[MB/sec]   MsgRate[Mpps]
     65536      1000             0.00               19827.40                   0.317238
    ---------------------------------------------------------------------------------------
    ************************************
    * Waiting for client to connect... * *
    ************************************
    ---------------------------------------------------------------------------------------
                        Send BW Test
     Dual-port       : OFF		Device         : mlx5_0
     Number of qps   : 1		Transport type : IB
     Connection type : RC		Using SRQ      : OFF
     PCIe relax order: ON
     ibv_wr* API     : ON
     TX depth        : 128
     CQ Moderation   : 100
     Mtu             : 4096[B]
     Link type       : IB
     Max inline data : 0[B]
     rdma_cm QPs	 : OFF
     Data ex.  method : Ethernet
    ---------------------------------------------------------------------------------------
     local address: LID 0x0f QPN 0x6d95 PSN 0xb974a
     remote address: LID 0x01 QPN 0x6dd2 PSN 0xc8a18c
    ---------------------------------------------------------------------------------------
     #bytes     #iterations    BW peak[MB/sec]    BW average[MB/sec]   MsgRate[Mpps]
     65536      1000             0.00               19827.40                   0.317238
    ---------------------------------------------------------------------------------------
    Example of mutual communication status check result

  3. Use the CLIENT Side command to check the communication status.

    Color mode
    ~$ ib_send_bw -d mlx5_0 -i 1 -F <SERVER Side IP>
    ~$ ib_send_bw -d mlx5_0 -i 1 -F <SERVER Side IP>
    CLIENT Side command example
    Color mode
    ---------------------------------------------------------------------------------------
                        Send BW Test
     Dual-port       : OFF		Device         : mlx5_0
     Number of qps   : 1		Transport type : IB
     Connection type : RC		Using SRQ      : OFF
     PCIe relax order: ON
     ibv_wr* API     : ON
     RX depth        : 512
     CQ Moderation   : 100
     Mtu             : 4096[B]
     Link type       : IB
     Max inline data : 0[B]
     rdma_cm QPs	 : OFF
     Data ex.  method : Ethernet
    ---------------------------------------------------------------------------------------
     local address: LID 0x01 QPN 0x6dd2 PSN 0xc8a18c
     remote address: LID 0x0f QPN 0x6d95 PSN 0xb974a
    ---------------------------------------------------------------------------------------
     #bytes     #iterations    BW peak[MB/sec]    BW average[MB/sec]   MsgRate[Mpps]
     65536      1000             19008.49            19006.37                  0.304102
    ---------------------------------------------------------------------------------------
    ---------------------------------------------------------------------------------------
                        Send BW Test
     Dual-port       : OFF		Device         : mlx5_0
     Number of qps   : 1		Transport type : IB
     Connection type : RC		Using SRQ      : OFF
     PCIe relax order: ON
     ibv_wr* API     : ON
     RX depth        : 512
     CQ Moderation   : 100
     Mtu             : 4096[B]
     Link type       : IB
     Max inline data : 0[B]
     rdma_cm QPs	 : OFF
     Data ex.  method : Ethernet
    ---------------------------------------------------------------------------------------
     local address: LID 0x01 QPN 0x6dd2 PSN 0xc8a18c
     remote address: LID 0x0f QPN 0x6d95 PSN 0xb974a
    ---------------------------------------------------------------------------------------
     #bytes     #iterations    BW peak[MB/sec]    BW average[MB/sec]   MsgRate[Mpps]
     65536      1000             19008.49            19006.37                  0.304102
    ---------------------------------------------------------------------------------------
    Example of mutual communication status check result

Check IB service related kernel modules

Check the relevant kernel modules for the IB service (lsmod) to inspect the IaaS hardware level.

  • Check IB service-related kernel module - nvidia_peermem

    Color mode
    ~$ lsmod | grep nvidia_peermem
    ~$ lsmod | grep nvidia_peermem
    Example command to check nvidia_peermem
    Color mode
    nvidia_peermem         16384  0
    ib_core               425984  9 rdma_cm,ib_ipoib,nvidia_peermem,iw_cm,ib_umad,rdma_ucm,ib_uverbs,mlx5_ib,ib_cm
    nvidia              56524800  451 nvidia_uvm,nvidia_peermem,nvidia_modeset
    nvidia_peermem         16384  0
    ib_core               425984  9 rdma_cm,ib_ipoib,nvidia_peermem,iw_cm,ib_umad,rdma_ucm,ib_uverbs,mlx5_ib,ib_cm
    nvidia              56524800  451 nvidia_uvm,nvidia_peermem,nvidia_modeset
    Example of nvidia_peermem check result

  • Check IB service related kernel modules - IB Card (HCA) driver

    Color mode
    ~$ lsmod|egrep 'ib_|_ib|mlx'
    ~$ lsmod|egrep 'ib_|_ib|mlx'
    Example command to check the IB Card (HCA) driver
    Color mode
    mlx5_ib               393216  0
    ib_uverbs             163840  2 irdma,mlx5_ib
    ib_core               393216  3 irdma,ib_uverbs,mlx5_ib
    mlx5_core            1593344  1 mlx5_ib
    mlxfw                  32768  1 mlx5_core
    psample                20480  1 mlx5_core
    tls                   114688  1 mlx5_core
    pci_hyperv_intf        16384  1 mlx5_core
    ib_ipoib              139264  0
    ib_cm                 131072  2 rdma_cm,ib_ipoib
    ib_umad                40960  0
    mlx5_ib               454656  0
    ib_uverbs             135168  2 rdma_ucm,mlx5_ib
    ib_core               434176  9 rdma_cm,ib_ipoib,nvidia_peermem,iw_cm,ib_umad,rdma_ucm,ib_uverbs,mlx5_ib,ib_cm
    libcrc32c              16384  5 nf_conntrack,nf_nat,btrfs,nf_tables,raid456
    mlx5_core            2064384  1 mlx5_ib
    mlx_compat             69632  11 rdma_cm,ib_ipoib,mlxdevm,iw_cm,ib_umad,ib_core,rdma_ucm,ib_uverbs,mlx5_ib,ib_cm,mlx5_core
    mlx5_ib               393216  0
    ib_uverbs             163840  2 irdma,mlx5_ib
    ib_core               393216  3 irdma,ib_uverbs,mlx5_ib
    mlx5_core            1593344  1 mlx5_ib
    mlxfw                  32768  1 mlx5_core
    psample                20480  1 mlx5_core
    tls                   114688  1 mlx5_core
    pci_hyperv_intf        16384  1 mlx5_core
    ib_ipoib              139264  0
    ib_cm                 131072  2 rdma_cm,ib_ipoib
    ib_umad                40960  0
    mlx5_ib               454656  0
    ib_uverbs             135168  2 rdma_ucm,mlx5_ib
    ib_core               434176  9 rdma_cm,ib_ipoib,nvidia_peermem,iw_cm,ib_umad,rdma_ucm,ib_uverbs,mlx5_ib,ib_cm
    libcrc32c              16384  5 nf_conntrack,nf_nat,btrfs,nf_tables,raid456
    mlx5_core            2064384  1 mlx5_ib
    mlx_compat             69632  11 rdma_cm,ib_ipoib,mlxdevm,iw_cm,ib_umad,ib_core,rdma_ucm,ib_uverbs,mlx5_ib,ib_cm,mlx5_core
    Example of IB Card (HCA) driver verification result

Check storage physical disk resources and Multi-Path

Verify the storage physical disk resources and Multi-Path to assess the IaaS hardware level.

  • Storage Physical Disk Resource Check Results

    Color mode
    root@bm-dev-002:/tmp# fdisk –l
    root@bm-dev-002:/tmp# fdisk –l
    Example of storage physical disk resource check result

  • Multi-Path verification result

    Color mode
    root@bm-dev-002:/tmp# multipath –ll
    root@bm-dev-002:/tmp# multipath –ll
    Example of Multi-Path verification result

Check Service Network after new deployment of Multi-node GPU Cluster

Use the following command to verify that the MII Status of Bonding and Slave Interface is up.

  • Service Network check command

    Color mode
    ~$ cat /proc/net/bonding/bond-srv
    ~$ cat /proc/net/bonding/bond-srv
    Example command to check Service Network

  • Service Network check result

    Color mode
    Ethernet Channel Bonding Driver: v5.15.0-25-generic
    Bonding Mode: fault-tolerance (active-backup)
    Primary Slave: None
    Currently Active Slave: ens9f0
    MII Status: up
    MII Polling Interval (ms): 100
    Up Delay (ms): 0
    Down Delay (ms): 0
    Peer Notification Delay (ms): 0
    
    Slave Interface: ens9f0
    MII Status: up
    Speed: 100000 Mbps
    Duplex: full
    Link Failure Count: 0
    Permanent HW addr: 30:3e:a7:02:35:70
    Slave queue ID: 0
    
    Slave Interface: ens11f0
    MII Status: up
    Speed: 100000 Mbps
    Duplex: full
    Link Failure Count: 0
    Permanent HW addr: 30:3e:a7:02:2f:e8
    Slave queue ID: 0
    Ethernet Channel Bonding Driver: v5.15.0-25-generic
    Bonding Mode: fault-tolerance (active-backup)
    Primary Slave: None
    Currently Active Slave: ens9f0
    MII Status: up
    MII Polling Interval (ms): 100
    Up Delay (ms): 0
    Down Delay (ms): 0
    Peer Notification Delay (ms): 0
    
    Slave Interface: ens9f0
    MII Status: up
    Speed: 100000 Mbps
    Duplex: full
    Link Failure Count: 0
    Permanent HW addr: 30:3e:a7:02:35:70
    Slave queue ID: 0
    
    Slave Interface: ens11f0
    MII Status: up
    Speed: 100000 Mbps
    Duplex: full
    Link Failure Count: 0
    Permanent HW addr: 30:3e:a7:02:2f:e8
    Slave queue ID: 0
    Example of Service Network verification result

Reference
If some Slave Interface is in a down state, use the Contact of the Support Center to report the abnormal situation and receive a response.

Multi-node GPU Cluster: Verify time synchronization with the Time Server after new deployment

The OS image includes the installation of the chrony daemon and configuration for SCP NTP server synchronization. Use the following command to verify whether a line marked with ^* exists in the MS Name column.

  • Command to check the chrony daemon source and synchronization status

    Color mode
    ~$ chronyc sources -V
    ~$ chronyc sources -V
    Example of status check command

  • Result of checking chrony daemon status

    Color mode
    MS Name/IP address         Stratum Poll Reach LastRx Last sample
    ===============================================================================
    ^* 198.19.0.54                   4  10   377  1040    -16us[  -37us] +/- 9982us
    MS Name/IP address         Stratum Poll Reach LastRx Last sample
    ===============================================================================
    ^* 198.19.0.54                   4  10   377  1040    -16us[  -37us] +/- 9982us
    Example of checking the chrony daemon status

5.3 - Release Note

Multi-node GPU Cluster

2026.03.19
FEATURE New OS version, server type addition and Terraform integration provided
  • Ubuntu image version 24.04 has been added.
  • B300 GPU server type has been added.
  • Provides an IaC environment through Terraform.
2025.07.01
FEATURE New feature addition and monitoring integration
  • You can terminate multiple resources simultaneously from the GPU Node list.
    • It must be a node that uses the same DataSet and Cluster Fabric.
  • Integrated with Cloud Monitoring.
    • You can view key performance metrics in real time in Cloud Monitoring.
2025.02.27
NEW Multi-node GPU Cluster official version launch
  • Multi-node GPU Cluster service has been launched.
    • We provide a service that offers physical GPU servers without virtualization for large-scale high-performance AI computing.

6 - Cloud Functions

6.1 - Overview

Service Overview

Cloud Functions is a serverless computing-based FaaS (Function as a Service) that lets you easily run function‑based applications without provisioning servers. Users don’t have to hassle with managing servers or containers for scaling, and can focus on writing code and deploying applications.

Features

  • Easy and Convenient Development Environment: Developers can easily create Function resources that connect to events across multiple environments using a Code Editor suitable for the selected runtime, and can write and invoke code with ease.
  • Serverless Computing: You can use a serverless code execution service for development in the Samsung Cloud Platform environment. * The resources required to invoke and run function-based applications are allocated and managed by the Samsung Cloud Platform according to the scale of execution.
  • Efficient Cost Management: The invoked Function aggregates usage (total number of calls, total execution time) and is billed for the time actually used to run the application. * Functions with low usage can have the Cloud Functions scaler adjusted to a scale-to-zero state, so they do not consume resources, enabling efficient cost management.

Service diagram

Diagram
Figure. Cloud_Functions diagram

Provided Features

Cloud Functions provides the following features.

  • Code Development Environment: Create Runtime-optimized functions, write and edit code
  • Function execution, environment management, monitoring: endpoint definition, Token management, access control configuration, trigger configuration, etc., runtime environment/variable definition and modification, artifact invocation/testing for Deploy/Test, service deployment, progress status monitoring/logging
  • Serverless Computing: All elements required for code writing and deployment are managed by Samsung Cloud Platform, with automatic scaling adjustments based on deployment.
  • Sample Code Provided: By providing various sample codes through Blueprint, you can start easily and quickly

Component

Runtime

Cloud Functions currently supports the following runtimes. Additional continuously supported runtimes will also be added.

RuntimeVersionDeprecation date
GO1.212026.7.30
GO1.232026.7.30
GO1.25-
java17-
Node.js182026.7.30
Node.js202026.7.30
Node.js22-
Node.js24-
PHP8.12026.7.30
PHP8.4-
PHP8.5-
Python3.92026.7.30
Python3.10-
Python3.11-
Python3.14-
Table. Supported Runtime Items
Caution

Runtime scheduled for deprecation

  • For Runtime versions slated for deprecation, technical support is expected to end within 60 days (End of Technical Support) (detailed schedule will be announced on the console).
  • When technical support for a Runtime version ends, security patches and updates will no longer be applied to that Runtime version.
    • Already created functions can be used continuously without a call time limit, but they cannot be modified.
    • We do not guarantee bugs, errors, defects, or vulnerabilities that arise in Runtime versions for which technical support has ended.
  • We recommend migrating to and using an alternative version of a function that is scheduled for deprecation to ensure safe usage.

Provision status by region

The Cloud Functions service is available in the environments below.

RegionWhether provided
Korea West 1 (kr-west1)Provide
Korea East 1 (kr-east1)Provided
South Korea 1 (kr-south1)Not provided
South Korea South 2 (kr-south2)Not provided
South Korea 3 (kr-south3)Not provided
Table. Cloud Functions regional availability status

Preceding Service

This is a list of services that can be configured as optional before creating the service. Please refer to the guide provided for each service and prepare in advance.

Service CategoryserviceDetailed description
Application ServiceAPI GatewayA service that easily manages and monitors APIs
Table. Cloud Functions prerequisite service

6.1.1 - ServiceWatch Metrics

Cloud Functions sends metrics to ServiceWatch. The metrics provided by default monitoring are data collected at a 1‑minute interval.

Reference
For how to view metrics in ServiceWatch, refer to the ServiceWatch guide.

Basic Metrics

The following are the basic metrics for the Cloud Functions namespace.

The metrics whose names are displayed in bold below are the metrics selected as key metrics among the default metrics provided by Cloud Functions. The key metrics are used to build service dashboards that are automatically created per service in ServiceWatch. You can also view the key metrics on the Monitoring tab of the Cloud Functions detail page.

Each metric guides users, through the user guide, on which statistical values are meaningful when querying that metric, and among the meaningful statistics, the values displayed in bold are the primary statistics. In the service dashboard or monitoring tab, you can view key metrics using these primary statistics.

Performance itemsDetailed descriptionunitmeaningful statistics
InvocationsAverage number of times the function is called per unit timeCount
  • Total
  • Average
  • Maximum
  • Minimum
Success CallsAverage number of times the runtime code operates correctly and returns a response code per unit time during a function call.Count
  • Total
  • Average
  • Maximum
  • Minimum
Error CallsAverage number of calls per unit time that encounter errors during function invocation, including runtime errors due to response timeouts and logic errors.Count
  • Total
  • Average
  • Maximum
  • Minimum
Memory UsageAverage memory usage per unit time while the function is executingKilobytes
  • Total
  • Average
  • Maximum
  • Minimum
Active OperationsWhen a function is called multiple times simultaneously, the average number of tasks generated per unit time for concurrent processing.Count
  • Total
  • Average
  • Maximum
  • Minimum
Table. Cloud Functions Basic Metrics

6.2 - How-to guides

Users can create the service by entering the required information for Cloud Functions and selecting detailed options through the Samsung Cloud Platform Console.

Creating Cloud Functions

  1. All Services > Compute > Cloud Functions Click the menu. 1. Go to the Service Home page of Cloud Functions.

  2. On the Service Home page, click the Create Cloud Functions button. 2. Go to the Create Cloud Functions page.

  3. On the Create Cloud Functions page, enter the information required to create the service.

    Category
    Required
    Detailed description
    Function nameRequiredEnter the name of the Funtion to create
    • Start with a lowercase English letter and use lowercase English letters, numbers, and special characters (-) to enter between 3 and 64 characters
    RuntimeRequiredSelect Runtime creation method
    • New: Create a new Runtime
    • Start with Blueprint: Write using the Runtime source code provided by the service
    Runtime & VesionEssentialSelect Runtime and Version
    • When Create New is selected
      • For the Java runtime, UI code editing is not supported, but you can import a JAR file from Object Storage and execute it
    • When Start with Blueprint is selected
      • You can view a source code example by clicking the View Source Code button for that Runtime & Version
    • If the Runtime version has reached End of Technical Support (EoTS), it cannot be modified after creation
    Table. Cloud Functions service information input fields

  4. Summary Check the detailed information and estimated charges generated in the panel, and click the Create button.

    • When creation is complete, check the created resources on the Cloud Functions list page.
information
After July 2026, you cannot create new functions for runtimes that are no longer supported. Note that already created user functions are not deleted.

View Cloud Functions details

Cloud Functions Details page consists of Details, Monitoring, Logs, Code, Configuration, Triggers, Tags, Job History tabs.

To view detailed information about the Cloud Functions service, follow these steps.

  1. All Services > Compute > Cloud Functions Click the menu. 1. Navigate to the Service Home page of Cloud Functions.
  2. On the Service Home page, click the Function menu. 2. Function list Go to the page.
  3. On the Function list page, click the resource to view detailed information. 3. Navigate to the Function Details page.
    • Function Details page displays status information and additional feature information, and consists of Details, Monitoring, Logs, Code, Configuration, Triggers, Tags, Job History tabs.
      CategoryDetailed description
      Cloud Functions statusCloud Functions status information
      • Ready: green icon, a state where normal function invocations are possible
      • Not Ready: gray icon, a state where normal function invocations are not possible
      • Deploying: yellow icon, a state where the function is being created or updated, which triggers the next action
        • Function creation and modification
        • Code tab: edit code in the editor
        • Code tab: inspect jar file
        • Trigger tab: add and modify
        • Configuration tab: modify
      • Running: blue icon, a state where normal function invocations are possible and a cold‑start prevention policy is applied
      Service cancellationCancel service button
      Table. Cloud Functions status information and additional features

Detailed Information

Function list page allows you to view detailed information of the selected resource and edit the information if needed.

CategoryDetailed description
serviceService name
Resource TypeResource Type
SRNUnique resource ID in Samsung Cloud Platform
Resource nameResource name
  • In the Cloud Functions service, it means the Function name
Resource IDService’s unique resource ID
ConstructorUser who created the service
Creation date and timeService creation timestamp
ModifierUser who modified the service
Modification dateDate and time of service modification
Function nameName of the Cloud Function
RuntimeRuntime types and versions
  • If the Runtime’s End of Technical Support (EoTS) is scheduled, display in yellow with a warning icon
Table. Cloud Functions Details – Details Tab Items
Caution
  • Even after technical support for the Runtime version ends, functions that have already been created can continue to be used without any call time limit. * However, security patches and updates for that Runtime version will not be applied.
    • Bugs, errors, defects, or vulnerabilities that occur in Runtime versions that are no longer supported are not covered.
  • If support for a Runtime has ended, the user must create a replacement Runtime version and then manually delete the function of the previous version.
    • To use this function safely, create a new instance using the Lastest or Stable version.

Monitoring

Function List page lets you view the Cloud Functions usage information for the selected resource.

CategoryDetailed description
Number of callsAverage number of times the function is called per unit time (instances)
execution timeAverage execution time (seconds) of the function per unit time
Memory usageAverage memory usage (KB) during the function execution per unit time
Current number of tasksWhen the function is called multiple times simultaneously, the average number of tasks (count) generated per unit time for concurrent processing.
Successful call countAverage number of times (cases) the runtime code operated correctly and returned a response code per unit time during a function call.
Failed call countAverage number of calls with errors per unit time during function invocation
  • Including runtime due to response timeouts and logic errors
Table. Cloud Functions Details – Monitoring Tab Items

log

Function list page allows you to view the Cloud Functions logs of the selected resource.

CategoryDetailed description
unit periodSelect the period to view Cloud Functions log information
  • Select in time units (1 hour, 3 hours, 12 hours) or allow the user to set a custom range
log messageFunctions are displayed in order, starting with the most recent occurrence.
Table. Cloud Functions Details – Log Tab Items
Reference
Log messages can be viewed up to the previous 1,000 entries based on the most recent occurrence.

code

Function List page lets you view and edit the Cloud Functions code of the selected resource.

Reference

The way to view and edit source code varies depending on the runtime used.

  • Inline Editor: Node.js, Python, PHP, Go
  • Compressed file (.jar/.zip) execution: Java
CategoryDetailed description
source codeInline editor method
code informationDisplay code information
EditAfter clicking the Edit button, you can modify the code in the inline editor.
Table. Cloud Functions Details – Inline Editor Items in the Code Tab
CategoryDetailed description
source codeExecution method for compressed files (.jar/.zip)
code informationDisplay compressed file information
  • Java Runtime: Java Runtime version information
  • Handler information: Execution class and method information
  • Compressed file name (.jar/.zip): Name of the currently configured compressed file
  • File upload timestamp: Upload timestamp of the currently configured compressed file
  • Transmission status: Compressed file transmission history
    • Transmission succeeded: When the compressed file configuration succeeds
    • Reason for failure when compressed file transmission fails
EditJar file can be modified
  • Cannot be modified after the Runtime version reaches end of support
  • On the Function code edit page, you can modify by clicking the Import from Object Storage button
  • Enter the Private URL of the file in the Object Storage bucket to be imported
Table. Cloud Functions Details - Execution items for compressed files (.jar/.zip) in the Code tab
Reference
  • If technical support for the runtime version has ended, you cannot modify the code. * Also, because security patches and updates are not applied, create and use the function anew with the Latest or Stable version to ensure safe usage.
  • In the case of Java Runtime, it does not provide a UI code editing feature, and you must select a compressed file (.jar/.zip) from a bucket in the Object Storage service.
  • If a user does not have an authentication key generated for the Object Storage service, they cannot execute Import from Object Storage, so they must create an authentication key in advance.
  • The Object Storage bucket for the Cloud Functions service must have its access control set to allow.

Configuration

On the Function list page, you can view the Cloud Functions configuration of the selected resource.

CategoryDetailed description
General configurationMemory and timeout settings of Cloud Function
  • Memory: Maximum memory limit that can be used per function
  • Timeout: Maximum time to wait for a function invocation per function
  • Function execution: Minimum and maximum number of tasks
  • Click the Edit button to modify the General configuration settings
function URLIssue an HTTPS URL address that can access the function
  • Enabled: Whether the function URL is enabled
  • Status: Current state of the function URL
    • Active: In use
    • Inactive: Disabled
    • Creating: Creating
  • Function URL: Click the function URL to navigate
  • Authentication type: When IAM is configured, only authenticated IAM users can access the function URL
  • Access control: When enabled, you can register and manage allowed IP addresses
  • Click the Edit button to configure Enabled status, Authentication type, and Allowed IPs
  • Refer to the OpenAPI of Samsung Cloud Platform
environment variableSet runtime environment variables
  • Environment variable: When used, you can adjust the function’s behavior without updating code
  • Edit button to add or modify environment variable
Private connection configurationCan be used in conjunction with PrivateLink Service
PermissionAdd and manage resource policies for IAM-based functions
  • Click the Edit button to edit the policy
  • Add: If there is no existing resource policy, add a new resource policy
  • Resource policies can be loaded from provided templates or created manually
Table. Cloud Functions Details - Configuration Tab Items
Caution
  • If technical support for the Runtime version has been discontinued, configuration items cannot be modified. * Also, because security patches and updates are not applied, create and use the function anew with the Latest or Stable version to ensure safe usage.
  • If access control is disabled, the registered access information is deleted, making function access control impossible, which can expose the system to security attacks such as external scanning, hacking, etc.
Reference
  • CPU cores proportional to the memory allocation of General configuration are automatically assigned.
  • If the minimum number of executions of General configuration is 1 or more, Cold Start is prevented, but continuous costs are incurred.

Trigger

On the Function List page, you can view and configure the trigger information of the selected resource. By setting a trigger, you can automatically execute the Function when an event occurs.

CategoryDetailed description
CronjobUse Cronjob as a trigger
  • Automatically invoke the function based on time or a scheduled interval
  • Edit button can be clicked to change repeat frequency and time zone
API GatewayUse API Gateway as a trigger
  • You can view the API Gateway name and detailed information
Table. Cloud Functions Details – Trigger Tab Items
Caution
  • If technical support for the Runtime version has ended, you cannot modify the trigger item. * Also, because security patches and updates are not applied, create and use the function anew with the Latest or Stable version to ensure safe usage.
  • If the Cronjob trigger is called before the function’s timeout, the function will execute concurrently, increasing both the execution count and the total time. * Therefore, be cautious because continuous additional costs can lead to high expenses.
Reference
  • If the status is Deploying, it cannot be modified.
  • Refer to Setting up triggers for trigger configuration.

Tag

In the Tag tab, you can view the resource’s tag information, and add, modify, or delete it.

CategoryDetailed description
Tag listTag list
  • You can view the Key, Value information of the tag
  • Up to 50 tags can be added per resource
  • When entering tags, search and select from the list of previously created Keys and Values
Table. Cloud Functions Details – Tag Tab Items

Job History

Job History page allows you to view the resource’s job history.

CategoryDetailed description
Task History ListResource Change History
  • You can view operation details, operation time, resource type, resource name, operation result, and operator information
  • Operation History List When you click the corresponding resource in the list, the Operation History Details popup opens
Table. Cloud Functions Details – Job History Tab Items

Changing Java Runtime code

If you are using Java Runtime, you cannot modify the code directly, so you must select and replace the archive file (.jar/.zip) in the bucket of the Object Storage service.

Reference
If technical support for the runtime version has ended, you cannot modify the code. Also, because security patches and updates are not applied, create and use the function anew with the Latest or Stable version to ensure safe usage.

To modify a compressed file, follow these steps.

  1. All Services > Compute > Cloud Functions menu, click. 1. Navigate to the Service Home page of Cloud Functions.

  2. On the Service Home page, click the Function menu. 2. Go to the Function list page.

  3. On the Function List page, click the resource to change the compressed file in the code. 3. Go to the Function Details page.

  4. Click the Edit button on the Code tab of the Function Details page. 4. Edit Function code Navigate to the page.

  5. Click the Import from Object Storage button. 5. Import from Object Storage The popup window opens.

    CategoryDetailed description
    Java RuntimeJava Runtime Information
    Handler informationHandler information
    • Execution Class: Automatically entered when setting the archive file (.jar/.zip)
    • Execution Method: Automatically entered when setting the archive file (.jar/.zip)
    Compressed file (.jar/.zip)Set the archive file to modify
    • Archive file name (.jar/.zip): Displays the name of the archive file. Import from Object Storage after configuration, it is entered automatically
    • Import from Object Storage: Configure the Object Storage to retrieve the archive file (.jar/.zip)
    Table. Cloud Functions Details - Function Code Modification Items

  6. Enter the URL information of the Object Storage to retrieve the compressed file in Object Storage URL, then click the Confirm button. 6. The notification popup opens.

    • URL information can be found in the Folder List tab of the detailed page of the Object Storage to retrieve, under the File Information > Private URL item.
  7. Click the Confirm button. 7. On the Function code edit page, the name of the imported compressed file is displayed in the Compressed file name (.jar/.zip).

  8. Click the Save button.

Caution
  • Users without a generated authentication key cannot execute Import from Object Storage.
  • If the URL does not exist or the archive file matches any of the following, it cannot be changed.
    • When using an unsupported file extension
    • If there are harmful files inside the compressed file.
    • If the size exceeds the supported limit

Terminate Cloud Functions

To cancel the Cloud Functions service, follow these steps.

  1. Click the All Services > Compute > Cloud Functions menu. 1. Navigate to the Service Home page of Cloud Functions.
  2. On the Service Home page, click the Function menu. 2. Navigate to the Function list page.
  3. On the Function List page, click the resource you want to terminate and then click the Terminate Service button.
  4. When the termination is complete, check on the Function list page whether the resource has been terminated.

6.2.1 - Configure Trigger

Configure Trigger

Note
  • By default, all triggers can be added in Cloud Functions.
  • If it is triggered for a specific product, it should be passed to Cloud Functions.

Setting up Cronjob trigger

To set up a Cronjob trigger, follow these steps.

  1. Click the All Services > Compute > Cloud Functions menu. Navigate to the Service Home page of Cloud Functions.
  2. On the Service Home page, click the Function menu. You will be taken to the Function list page.
  3. Function List page, click the resource for which you want to set a trigger. You will be taken to the Function Details page.
  4. After clicking the Trigger tab, click the Add Trigger button. Set it. The Add Trigger popup window opens.
  5. Add Trigger In the popup, select Cronjob from Trigger Type. A required information input area appears at the bottom.
    CategoryDetailed description
    Cronjob configurationSet the trigger’s repeat frequency
    • Can be set in minutes, hours, days, months, weekdays
    Timezone settingSet the trigger’s reference time zone
    Table. Cronjob Trigger Required Information Items
  6. After entering the required information, click the Confirm button.
  7. When the popup notifying an addition opens, click the Confirm button.

Configure API Gateway Trigger

To set up an API Gateway trigger, follow these steps.

  1. Click the All Services > Compute > Cloud Functions menu. Navigate to the Service Home page of Cloud Functions.
  2. On the Service Home page, click the Function menu. You will be taken to the Function List page.
  3. Click the resource to set the trigger on the Function List page. Go to the Function Details page.
  4. Click the Trigger tab, then click the Add Trigger button. Set it. The Add Trigger popup opens.
  5. In the Add Trigger popup, select API Gateway under Trigger Type. A required information input area appears at the bottom.
    CategoryDetailed description
    API nameSelect API
    • You can select an existing API or create a new one
    StageSelect deployment target
    • You can select an existing stage or create a new one
    Table. API Gateway Trigger Required Information Items
  6. After entering the required information, click the Confirm button.
  7. When the popup notifying the addition opens, click the Confirm button.

Configure Multi-Trigger

You can attach multiple triggers to a single function.

Modify Trigger

To modify the added trigger, follow these steps.

  1. Click the All Services > Compute > Cloud Functions menu. Navigate to the Service Home page of Cloud Functions.
  2. On the Service Home page, click the Function menu. You will be taken to the Function list page.
  3. On the Function List page, click the resource to edit the trigger. You will be taken to the Function Details page.
  4. Click the Trigger tab, then in the trigger list, click the Edit button of the trigger whose settings you want to modify. The Edit Trigger popup window opens.
  5. Edit Trigger After modifying the settings in the popup window, click the Confirm button.
  6. When the edit notification popup appears, click Confirm.

Delete Trigger

To delete a trigger, follow these steps.

Caution
A trigger linked to a specific product manages only the product delivered at the time of linking, and when the Functions are terminated, it must convey a deletion status to that product.
  1. Click the All Services > Compute > Cloud Functions menu. Navigate to the Service Home page of Cloud Functions.
  2. On the Service Home page, click the Function menu. You will be taken to the Function List page.
  3. Function List page, click the resource for which you want to set a trigger. You will be taken to the Function Details page.
  4. In the Trigger tab’s trigger list, select the trigger you want to delete, then click the Delete button.
  5. When the popup notifying you of trigger deletion opens, click the Confirm button.

6.2.2 - Blueprint Detailed Guide

Blueprint Overview

When creating Cloud Functions, you can set a Blueprint to utilize the Runtime source code provided by Cloud Functions. Refer to the following for the Blueprint items provided by Cloud Functions.

CategoryDetailed descriptionRemarks
Hello WorldWhen the function is invoked, it responds with Hello Serverless World!
Execution after timeoutIt outputs code that should run after the function call timeout but does not execute.PHP, Python not supported
HTTP request bodyParse the request body.PHP not supported
Send HTTP requestsThe Cloud function sends an HTTP request.PHP not supported
Print logsLogs the user’s Samsung Cloud Platform Console request.PHP not supported
Throw a custom errorEnter the error logic directly to handle the error.
Using Environment VariableConfigure environment variables within the Cloud function and execute it.
Table. Blueprint Items

Hello World

Hello World Explains the response-receiving configuration and a function call example (using the function URL).

Hello World Setup

To set up Hello World, follow these steps.

  1. Click the All Services > Compute > Cloud Functions menu. Go to the Service Home page of Cloud Functions.

  2. On the Service Home page, click the Function menu. You will be taken to the Function List page.

  3. Function List page, click the resource to be called via URL. You will be taken to the Function Detail page.

  4. After clicking the Configuration tab, click the Edit button for the Function URL item. The Edit Function URL popup window opens.

  5. In the Function URL Edit popup, set Activation status to Enabled, then click the Confirm button.

    CategoryDetailed description
    Enable statusConfigure the use of the function URL
    Authentication typeSelect whether to use IAM authentication for requests to the function URL
    Access controlAdd accessible IPs to enable management
    • Set to Use, then you can input and add a public access IP
    Table. Required input fields when adding a trigger

  6. After navigating to the Code tab, click the Edit button. You will be taken to the Function Code Edit page.

  7. After adding the handling logic for success and failure cases, click the Save button.

    • Node.js source code
      Color mode
      exports.handleRequest = async function (params) {
          /**
          * @description User writing area (Function details)
          */
          const response = {
          statusCode: 200,
          body: JSON.stringify('Hello Serverless World!'),
          };
          return response;
      };
      exports.handleRequest = async function (params) {
          /**
          * @description User writing area (Function details)
          */
          const response = {
          statusCode: 200,
          body: JSON.stringify('Hello Serverless World!'),
          };
          return response;
      };
      Hello World - Node.js source code
    • Python source code
      Color mode
      import json
      
      def handle_request(params):
          # User writing area (Function details)
          return {
          'statusCode': 200,
          'body': json.dumps('Hello Serverless World!')
          }
      import json
      
      def handle_request(params):
          # User writing area (Function details)
          return {
          'statusCode': 200,
          'body': json.dumps('Hello Serverless World!')
          }
      Hello World - Python source code
    • PHP source code
      Color mode
      <?php
      function handle_request() {
          # User writing area (Function details)
          $res = array(
              'statusCode' => 200,
              'body' => 'Hello Serverless World!',
          );
          return $res;
      }
      ?>
      
      <?php
      function handle_request() {
          # User writing area (Function details)
          $res = array(
              'statusCode' => 200,
              'body' => 'Hello Serverless World!',
          );
          return $res;
      }
      ?>
      
      Hello World - PHP source code

Check function call

On the Function Details page, in the Configuration tab, invoke the function URL and then verify the response.

Hello Serverless World!

Execution after timeout

Describes configuring execution after timeout (Execution after timeout) and provides an example of invoking the function (using the function URL).

Configure execution after timeout

To set Execution after timeout, follow these steps.

  1. Click the All Services > Compute > Cloud Functions menu. Navigate to the Service Home page of Cloud Functions.
  2. On the Service Home page, click the Function menu. You will be taken to the Function List page.
  3. On the Function List page, click the resource for which you want to set a trigger. You will be taken to the Function Details page.
  4. After clicking the Trigger tab, click the Add Trigger button. The Add Trigger popup window opens.
  5. Add Trigger In the popup window, after selecting the Trigger Type item, enter the required information displayed at the bottom and click the OK button.
    • Required information varies depending on the trigger type.
      Trigger TypesInput field
      API Gateway
      • API name: Select an existing API or create a new one
      • Stage: Select an existing stage or create a new one
      Cronjob
      • Refer to the example and enter the trigger’s repeat frequency(minute, hour, day, month, day of week)
      • Timezone setting: select the reference time zone to apply
      Table. Required input fields when adding a trigger
  6. After moving to the Code tab, click the Edit button. You will be taken to the Function Code Edit page.
  7. After adding the handling logic for success and failure cases, click the Save button.
    • Node.js source code
      Color mode
      exports.handleRequest = async function (params) {
          /**
           * @description User writing area (Function details)
           */
          console.log("Hello world 3");
          await delay(3000);
      
          const response = {
              statusCode: 200,
              body: JSON.stringify('Hello Serverless World!'),
          };
          return response;
      };
      
      const delay = (ms) => {
          return new Promise(resolve=>{
              setTimeout(resolve,ms)
          })
      }
      exports.handleRequest = async function (params) {
          /**
           * @description User writing area (Function details)
           */
          console.log("Hello world 3");
          await delay(3000);
      
          const response = {
              statusCode: 200,
              body: JSON.stringify('Hello Serverless World!'),
          };
          return response;
      };
      
      const delay = (ms) => {
          return new Promise(resolve=>{
              setTimeout(resolve,ms)
          })
      }
      Execution after timeout - Node.js source code

Check function call

On the Function Detail page’s Configuration tab, invoke the function URL and, after a brief period, check the response.

Hello Serverless World!

HTTP request body

Explains the configuration for parsing the Request Body and an example of calling the function (using the function URL).

Setting HTTP request body

To set the HTTP request body, follow these steps.

  1. Click the All Services > Compute > Cloud Functions menu. Navigate to the Service Home page of Cloud Functions.
  2. On the Service Home page, click the Function menu. You will be taken to the Function List page.
  3. Function List page, click the resource to set the trigger. You will be taken to the Function Details page.
  4. After clicking the Trigger tab, click the Add Trigger button. The Add Trigger popup window opens.
  5. Add Trigger In the popup window, select the Trigger Type option, then fill in the required information shown at the bottom and click the OK button.
    • Required information varies depending on the trigger type.
      Trigger TypesInput field
      API Gateway
      • API name: Select an existing API or create a new one
      • Stage: Select an existing stage or create a new one
      Cronjob
      • Refer to the example and enter the trigger’s repeat frequency (minutes, hours, day, month, day of week)
      • Timezone setting: select the reference time zone to apply
      Table. Required input fields when adding a trigger
  6. After moving to the Code tab, click the Edit button. You will be taken to the Function Code Edit page.
  7. After adding the handling logic for success and failure cases, click the Save button.
    • Node.js source code
      Color mode
      exports.handleRequest = async function (params) {
          /**
          * @description User writing area (Function details)
          */
          const response = {
          statusCode: 200,
          body: JSON.stringify(params.body),
          };
          return response;
      };
      exports.handleRequest = async function (params) {
          /**
          * @description User writing area (Function details)
          */
          const response = {
          statusCode: 200,
          body: JSON.stringify(params.body),
          };
          return response;
      };
      Execution after timeout - Node.js source code
    • Python source code
      Color mode
      import json
      
      def handle_request(params):
          # User writing area (Function details)
          return {
              'statusCode': 200,
              'body': json.dumps(params.json)
      }
      import json
      
      def handle_request(params):
          # User writing area (Function details)
          return {
              'statusCode': 200,
              'body': json.dumps(params.json)
      }
      Execution after timeout - Python source code

Check function call

In the Configuration tab of the Function Details page, after calling the Function URL, check the Body data, request Body value, and response Body value.

  • Request Body value

    Color mode
    {
        "testKey" :"cloud-001",
        "testNames": [
            {
                "name": "Son"
            },
            {
                "name": "Kim"
            }
        ],
        "testCode":"test"
    }
    {
        "testKey" :"cloud-001",
        "testNames": [
            {
                "name": "Son"
            },
            {
                "name": "Kim"
            }
        ],
        "testCode":"test"
    }
    Request Body value

  • Response Body value

    Color mode
    {
        "testKey" :"cloud-001",
        "testNames": [
            {
                "name": "Son"
            },
            {
                "name": "Kim"
            }
        ],
        "testCode":"test"
    }
    {
        "testKey" :"cloud-001",
        "testNames": [
            {
                "name": "Son"
            },
            {
                "name": "Kim"
            }
        ],
        "testCode":"test"
    }
    Response Body value

Send HTTP requests

Explains the HTTP request configuration and an example of calling a function (using the function URL).

Send HTTP requests Configure

To configure Send HTTP requests, follow these steps.

  1. Click the All Services > Compute > Cloud Functions menu. Navigate to the Service Home page of Cloud Functions.
  2. On the Service Home page, click the Function menu. You will be taken to the Function List page.
  3. Click the resource to set the trigger on the Function List page. Go to the Function Details page.
  4. After clicking the Trigger tab, click the Add Trigger button. The Add Trigger popup window opens.
  5. Add Trigger In the popup window, after selecting the Trigger Type item, enter the required information displayed at the bottom and click the OK button.
    • Required information varies depending on the trigger type.
      Trigger TypesInput field
      API Gateway
      • API name: Select an existing API or create a new one
      • Stage: Select an existing stage or create a new one
      Cronjob
      • Refer to the example and enter the trigger’s repeat frequency (minutes, hours, day, month, day of week)
      • Timezone setting: select the reference time zone to apply
      Table. Required input items when adding a trigger
  6. After moving to the Code tab, click the Edit button. You will be taken to the Function Code Edit page.
  7. After adding the handling logic for success and failure cases, click the Save button.
    • Node.js source code
      Color mode
      const request = require('request');
      
      /**
      * @description User writing area (Function details)
      */
      exports.handleRequest = async function (params) {
      return await sendRequest(params);
      };
      
      async function sendRequest(req) {
          return new Promise((resolve, reject) => {
              // Port 80 and Port 443 are available
              url = "https://example.com"; // Destination URL
      
                  const options = {
                  uri: url,
                  method:'GET',
                  json: true,
                  strictSSL: false,
                  rejectUnauthorized: false
              }
              request(options, (error, response, body) => {
                  if (error) {
                      reject(error);
                  } else {
                      resolve({
                          statusCode: response.statusCode,
                          body: JSON.stringify(body)
                      });
                  }
              });
          });
      }
      const request = require('request');
      
      /**
      * @description User writing area (Function details)
      */
      exports.handleRequest = async function (params) {
      return await sendRequest(params);
      };
      
      async function sendRequest(req) {
          return new Promise((resolve, reject) => {
              // Port 80 and Port 443 are available
              url = "https://example.com"; // Destination URL
      
                  const options = {
                  uri: url,
                  method:'GET',
                  json: true,
                  strictSSL: false,
                  rejectUnauthorized: false
              }
              request(options, (error, response, body) => {
                  if (error) {
                      reject(error);
                  } else {
                      resolve({
                          statusCode: response.statusCode,
                          body: JSON.stringify(body)
                      });
                  }
              });
          });
      }
      Send HTTP requests - Node.js source code
    • Python source code
      Color mode
      import json
      import requests
      
      def handle_request(params):
          # User writing area (Function details)
          
          # Port 80 and Port 443 are available
          url = "https://example.com" # Destination URL
      
          try:
              response = requests.get(url, verify=True)
              return {
                  'statusCode': response.status_code,
                  'body': json.dumps(response.text)
              }
          except requests.exceptions.RequestException as e:
              return str(e)
      import json
      import requests
      
      def handle_request(params):
          # User writing area (Function details)
          
          # Port 80 and Port 443 are available
          url = "https://example.com" # Destination URL
      
          try:
              response = requests.get(url, verify=True)
              return {
                  'statusCode': response.status_code,
                  'body': json.dumps(response.text)
              }
          except requests.exceptions.RequestException as e:
              return str(e)
      Send HTTP requests - Python source code

Check Function Call

On the Function Details page, after invoking the function URL in the Configuration tab, verify the response.

Color mode
<!doctype html>
<html>
<head>
    <title>Example Domain</title>

    <meta charset="utf-8" />
    <meta http-equiv="Content-type" content="text/html; charset=utf-8" />
    <meta name="viewport" content="width=device-width, initial-scale=1" />
    <style type="text/css">
    body {
        background-color: #f0f0f2;
        margin: 0;
        padding: 0;
        font-family: -apple-system, system-ui, BlinkMacSystemFont, "Segoe UI", "Open Sans", "Helvetica Neue", Helvetica, Arial, sans-serif;
    }
    div {
        width: 600px;
        margin: 5em auto;
        padding: 2em;
        background-color: #fdfdff;
        border-radius: 0.5em;
        box-shadow: 2px 3px 7px 2px rgba(0,0,0,0.02);
    }
    a:link, a:visited {
        color: #38488f;
        text-decoration: none;
    }
    @media (max-width: 700px) {
        div {
            margin: 0 auto;
            width: auto;
        }
    }
    </style>
</head>

<body>
<div>
    <h1>Example Domain</h1>

    <p>This domain is for use in illustrative examples in documents. You may use this
    domain in literature without prior coordination or asking for permission.</p>
    <p><a href="https://www.iana.org/domains/example">More information...</a></p>
</div>
</body>
</html>
<!doctype html>
<html>
<head>
    <title>Example Domain</title>

    <meta charset="utf-8" />
    <meta http-equiv="Content-type" content="text/html; charset=utf-8" />
    <meta name="viewport" content="width=device-width, initial-scale=1" />
    <style type="text/css">
    body {
        background-color: #f0f0f2;
        margin: 0;
        padding: 0;
        font-family: -apple-system, system-ui, BlinkMacSystemFont, "Segoe UI", "Open Sans", "Helvetica Neue", Helvetica, Arial, sans-serif;
    }
    div {
        width: 600px;
        margin: 5em auto;
        padding: 2em;
        background-color: #fdfdff;
        border-radius: 0.5em;
        box-shadow: 2px 3px 7px 2px rgba(0,0,0,0.02);
    }
    a:link, a:visited {
        color: #38488f;
        text-decoration: none;
    }
    @media (max-width: 700px) {
        div {
            margin: 0 auto;
            width: auto;
        }
    }
    </style>
</head>

<body>
<div>
    <h1>Example Domain</h1>

    <p>This domain is for use in illustrative examples in documents. You may use this
    domain in literature without prior coordination or asking for permission.</p>
    <p><a href="https://www.iana.org/domains/example">More information...</a></p>
</div>
</body>
</html>
Check function call response

Print logs

This explains how to configure log output and an example of calling a function (using the function URL).

Configure Print logs

Print logs To set up response receiving, follow these steps.

  1. Click the All Services > Compute > Cloud Functions menu. Navigate to the Service Home page of Cloud Functions.
  2. On the Service Home page, click the Function menu. You will be taken to the Function List page.
  3. Function List page, click the resource to set the trigger. Function Details page will be displayed.
  4. After clicking the Trigger tab, click the Add Trigger button. The Add Trigger popup window opens.
  5. In the Add Trigger popup, select the Trigger Type item, then enter the required information displayed at the bottom and click the OK button.
    • Required information varies depending on the trigger type.
      Trigger TypesInput field
      API Gateway
      • API name: Select an existing API or create a new one
      • Stage: Select an existing stage or create a new one
      Cronjob
      • Refer to the example and enter the trigger’s repeat frequency(minutes, hours, day, month, day of week)
      • Timezone setting: select the reference time zone to apply
      Table. Required input fields when adding a trigger
  6. After moving to the Code tab, click the Edit button. You will be taken to the Function Code Edit page.
  7. After adding the handling logic for success and failure cases, click the Save button.
    • Node.js source code
      Color mode
      const winston = require('winston');
      
      // Log module setting
      const logger = winston.createLogger({
          format: winston.format.combine(
              winston.format.timestamp(),
              winston.format.printf(info => info.timestamp + ' ' + info.level + ': ' + info.message)
              ),
              transports: [
                  new winston.transports.Console()
                  ]
      });
      
      exports.handleRequest = async function (params) {
          /**
          * @description User writing area (Function details)
          */
          const response = {
              statusCode: 200,
              body: JSON.stringify(params.body),
          };
      
          logger.info(JSON.stringify(response, null, 2));
      
          return response;
      };
      const winston = require('winston');
      
      // Log module setting
      const logger = winston.createLogger({
          format: winston.format.combine(
              winston.format.timestamp(),
              winston.format.printf(info => info.timestamp + ' ' + info.level + ': ' + info.message)
              ),
              transports: [
                  new winston.transports.Console()
                  ]
      });
      
      exports.handleRequest = async function (params) {
          /**
          * @description User writing area (Function details)
          */
          const response = {
              statusCode: 200,
              body: JSON.stringify(params.body),
          };
      
          logger.info(JSON.stringify(response, null, 2));
      
          return response;
      };
      Print logs - Node.js source code
    • Python source code
      Color mode
      import json
      import logging
      
      # Log module setting
      logging.basicConfig(level=logging.INFO)
      
      def handle_request(params):
          # User writing area (Function details)
          response = {
              'statusCode': 200,
              'body': json.dumps(params.json)
          }
      
          logging.info(response)
      
          return response
      import json
      import logging
      
      # Log module setting
      logging.basicConfig(level=logging.INFO)
      
      def handle_request(params):
          # User writing area (Function details)
          response = {
              'statusCode': 200,
              'body': json.dumps(params.json)
          }
      
          logging.info(response)
      
          return response
      Print logs - Python source code

Check Function Call

After calling the function URL in the Configuration tab of the Function Details page, check the log in the Log tab.

Color mode
[2023-09-07] 12:06:23] "host": "scf-xxxxxxxxxxxxxxxxxxxxx",
[2023-09-07] 12:06:23] "ce-id": "xxxxxxxxxxxxxxxxxxxxx",
[2023-09-07] 12:06:23] "ce-source": "xxxxxxxxxxxxxxxxxxxxx",
[2023-09-07] 12:06:23] "host": "scf-xxxxxxxxxxxxxxxxxxxxx",
[2023-09-07] 12:06:23] "ce-id": "xxxxxxxxxxxxxxxxxxxxx",
[2023-09-07] 12:06:23] "ce-source": "xxxxxxxxxxxxxxxxxxxxx",
Check function call response

Throw a custom error

Explains setting up a custom error (Throw a custom error) and an example of calling a function (using a function URL).

Configure Throw a custom error

To configure Throw a custom error, follow these steps.

  1. Click the All Services > Compute > Cloud Functions menu. Navigate to the Service Home page of Cloud Functions.
  2. On the Service Home page, click the Function menu. You will be taken to the Function List page.
  3. Click the resource to set the trigger on the Function List page. Go to the Function Details page.
  4. After clicking the Trigger tab, click the Add Trigger button. The Add Trigger popup window opens.
  5. Add Trigger In the popup window, after selecting the Trigger Type item, enter the required information displayed at the bottom and click the OK button.
    • Required information varies depending on the trigger type.
      Trigger TypesInput field
      API Gateway
      • API name: Select an existing API or create a new one
      • Stage: Select an existing stage or create a new one
      Cronjob
      • Refer to the example and enter the trigger’s repeat frequency (minutes, hours, day, month, day of week)
      • Timezone setting: select the reference time zone to apply
      Table. Required input fields when adding a trigger
  6. After moving to the Code tab, click the Edit button. You will be taken to the Function Code Edit page.
  7. After adding the handling logic for success and failure cases, click the Save button.
    • Node.js source code
      Color mode
      class CustomError extends Error {
          constructor(message) {
              super(message);
              this.name = 'CustomError';
          }
      }
      
      exports.handleRequest = async function (params) {
          /**
          * @description User writing area (Function details)
          */
          throw new CustomError('This is a custom error!');
      };
      class CustomError extends Error {
          constructor(message) {
              super(message);
              this.name = 'CustomError';
          }
      }
      
      exports.handleRequest = async function (params) {
          /**
          * @description User writing area (Function details)
          */
          throw new CustomError('This is a custom error!');
      };
      Throw a custom error - Node.js source code
    • Python source code
      Color mode
      class CustomError(Exception):
          def __init__(self, message):
          self.message = message
      
      def handle_request(parmas):
          raise CustomError('This is a custom error!')
      class CustomError(Exception):
          def __init__(self, message):
          self.message = message
      
      def handle_request(parmas):
          raise CustomError('This is a custom error!')
      Throw a custom error - Python source code
    • PHP source code
      Color mode
      <?php
          class CustomError extends Exception {
              public function __construct($message) {
                  parent::__construct($message);
                  $this->message = $message;
              }
          }
      
          function handle_request() {
              throw new CustomError('This is a custom error!');
          }
      ?>
      
      <?php
          class CustomError extends Exception {
              public function __construct($message) {
                  parent::__construct($message);
                  $this->message = $message;
              }
          }
      
          function handle_request() {
              throw new CustomError('This is a custom error!');
          }
      ?>
      
      Throw a custom error - PHP source code

Check Function Call

On the Function Details page, after calling the Function URL in the Configuration tab, verify whether an error occurred in the Log tab.

Using Environment Variable

Explains the use of environment variables (Using Environment Variable) settings and a function call example (using function URL).

Using Environment Variable Configure

To configure Using Environment Variable, follow these steps.

  1. Click the All Services > Compute > Cloud Functions menu. Navigate to the Service Home page of Cloud Functions.
  2. On the Service Home page, click the Function menu. You will be taken to the Function List page.
  3. Click the resource to set the trigger on the Function List page. Go to the Function Details page.
  4. After clicking the Trigger tab, click the Add Trigger button. The Add Trigger popup window opens.
  5. In the Add Trigger popup, select the Trigger Type item, then enter the required information displayed at the bottom and click the OK button.
    • Required information varies depending on the trigger type.
      Trigger TypesInput field
      API Gateway
      • API name: Select an existing API or create a new one
      • Stage: Select an existing stage or create a new one
      Cronjob
      • Refer to the example and enter the trigger’s repeat frequency(minutes, hours, days, months, day of week)
      • Timezone setting: select the reference time zone to apply
      Table. Required input fields when adding a trigger
  6. After moving to the Code tab, click the Edit button. You will be taken to the Function Code Edit page.
  7. After adding the handling logic for success and failure cases, click the Save button.
    • Node.js source code
      Color mode
      exports.handleRequest = async function (params) {
          /**
          * @description User writing area (Function details)
          */
          return process.env.test;
      };
      exports.handleRequest = async function (params) {
          /**
          * @description User writing area (Function details)
          */
          return process.env.test;
      };
      Using Environment Variable - Node.js source code
    • Python source code
      Color mode
      import json
      
      import os
      
      def handle_request(params):
          # User writing area (Function details)
          return os.environ.get("test")
      import json
      
      import os
      
      def handle_request(params):
          # User writing area (Function details)
          return os.environ.get("test")
      Using Environment Variable - Python source code
    • PHP source code
      Color mode
      import json
      
      def handle_request(params):
          # User writing area (Function details)
          return os.environ.get("test")
      import json
      
      def handle_request(params):
          # User writing area (Function details)
          return os.environ.get("test")
      Using Environment Variable - PHP source code
  8. After moving to the Configuration tab, click the Edit button in the Environment Variables area. The Edit Environment Variables popup will open.
  9. After entering the environment variable information, click the Confirm button.
    CategoryDetailed description
    NameEnter the key value
    valueEnter the value
    Table. Environment Variable Input Items

Check function call

On the Function Details page, after calling the function URL in the Configuration tab, check the environment variable value in the Log tab.

6.2.3 - Integrate PrivateLink Service

By integrating Cloud Functions with the PrivateLink service, you can connect VPCs within the Samsung Cloud Platform to other VPCs, and VPCs to services, without using the external internet.
The data uses only the internal network, providing high security, and does not require public IP, NAT, VPN, or an internet gateway.

Enable PrivateLink Service

To connect the PrivateLink Service, you must first enable the service.

To enable the PrivateLink service, follow these steps.

  1. All Services > Compute > Cloud Functions Click the menu. 1. Navigate to the Service Home page of Cloud Functions.
  2. On the Service Home page, click the Function menu. 2. Go to the Function list page.
  3. On the Function List page, click the resource to associate with PrivateLink. 3. Function Details page.
  4. On the Function Details page, click the Configuration tab.
  5. In Private connection configuration, click the Edit button of PrivateLink Service. 5. PrivateLink Service Edit The popup window opens.
  6. PrivateLink Service Edit In the popup window, after checking the Use item of Activation Status, click the Confirm button. 6. Configuration tab’s Private connection configuration displays PrivateLink Service information.
CategoryDetailed description
Private URLPrivateLink Service URL information
PrivateLink Service IDPrivateLink Service ID information
Request Endpoint ManagementList of PrivateLink Endpoints that requested a PrivateLink Service connection
  • Endpoint ID and approval status
  • Approval Management button can be clicked to change the status
    • Requesting: Endpoint that is requesting a connection. Click the Approve or Reject button to select approval
    • Active: Endpoint with a completed connection. Click the Block button to disconnect
    • Disconnected: Endpoint whose connection has been terminated. Click the Reconnect button to re-establish the connection
    • Reject: Endpoint whose connection request was denied
Table. PrivateLink Service detailed information items

Integrating PrivateLink Service

You can expose the function for private access from another VPC by integrating with PrivateLink Service.

information
Activate the PrivateLink Service first, then proceed with the integration work.

To integrate the PrivateLink service, review the following tasks.

  • Register the domain for the PrivateLink Endpoint IP address and the Private URL address to invoke the issued Private URL.
    192.168.0.13 abc123.scf.private.kr-west1.qa2.samsungsdscloud.com
    
  • When invoking the PrivateLink Service, verify IAM authentication based on the credentials of the Endpoint creator required for the Endpoint.

Create PrivateLink Endpoint

Create an entry point to access the PrivateLink Service of the user VPC.

Caution
Additional costs may be incurred when creating an endpoint.

To create a PrivateLink Endpoint, follow these steps.

  1. Click the All Services > Compute > Cloud Functions menu. 1. Go to the Service Home page of Cloud Functions.
  2. On the Service Home page, click the Function menu. 2. Go to the Function list page.
  3. On the Function list page, click the resource to associate with PrivateLink. 3. Function Details Go to the page.
  4. On the Function Details page, click the Configuration tab.
  5. Click the Add button in Private connection configuration of PrivateLink Endpoint. 5. Add PrivateLink Endpoint The popup window opens.
  6. Add PrivateLink Service in the popup window, after entering the PrivateLink Service ID and Alias information, click the Confirm button.
  7. When the popup indicating creation opens, click the Confirm button. 7. Configuration tab’s Private connection configuration displays PrivateLink Endpoint information.
CategoryDetailed description
PrivateLink Endpoint IDPrivateLink Endpoint ID information
PrivateLink Service IDPrivateLink Service ID information
Aliashostalias information that can be used instead of an IP address for accessing a PrivateLink Endpoint
statusApproval status of PrivateLink Endpoint
  • Requesting: Pending approval
  • Active: Approved and connected
  • Disconnected: Disconnected
  • Reject: Approval rejected. Click the Retry button to retry
  • Delete: Delete the endpoint
Table. PrivateLink Endpoint detailed information items

Integrating APIGW Private EPS

To connect the SCF Endpoint and the APIGW Private Endpoint, you must specify the Private URL in the SCF Endpoint Alias instead of the APIGW EPS resource path.

  • Private URL example: 181b6126ef6d4e4b81370df5.apigw.private.kr-west1.s.samsungsdscloud.com/get/resourcepath

To integrate APIGW Private EPS, refer to the following code.

Color mode
const request = require('request');

/**
 * @description User writing area (Function details)
*/
exports.handleRequest = async function (params) {
    return await sendRequest(params);
};
 
async function sendRequest(req) {
    return new Promise((resolve, reject) => {
        // Port 80 and Port 443 are available
        url = "https://{alias}/{resource_path}"; // Destination URL
        /**
        {alias} is the alias name entered when creating an Endpoint within the function
        {resoure_path} is the resource path (/get/resourcepath) specified in the Private URL of APIGW EPS
        */

        const options = {
            uri: url,
            method:'GET',
            json: true,
            strictSSL: false,
            rejectUnauthorized: false
        }
         
        request(options, (error, response, body) => {
            if (error) {
                reject(error);
            } else {
                resolve({
                    statusCode: response.statusCode,
                    body: JSON.stringify(body)
                });
            }
        });
    });
}
const request = require('request');

/**
 * @description User writing area (Function details)
*/
exports.handleRequest = async function (params) {
    return await sendRequest(params);
};
 
async function sendRequest(req) {
    return new Promise((resolve, reject) => {
        // Port 80 and Port 443 are available
        url = "https://{alias}/{resource_path}"; // Destination URL
        /**
        {alias} is the alias name entered when creating an Endpoint within the function
        {resoure_path} is the resource path (/get/resourcepath) specified in the Private URL of APIGW EPS
        */

        const options = {
            uri: url,
            method:'GET',
            json: true,
            strictSSL: false,
            rejectUnauthorized: false
        }
         
        request(options, (error, response, body) => {
            if (error) {
                reject(error);
            } else {
                resolve({
                    statusCode: response.statusCode,
                    body: JSON.stringify(body)
                });
            }
        });
    });
}
APIGW Private EPS integration code

6.2.4 - Resource-based Policy Guide

Resource-based policy overview

The resource-based policy (Resource-based Policy) of Cloud Functions is a policy granted to a resource that can decide to allow or deny (Effect) an action (Action) on a specific resource for a principal (Principal). You can directly define the principal that can invoke a function by using resource-based policies.

Reference
While a typical IAM policy (Identity-based) grants permissions to a user, a resource-based policy is applied to the function itself to allow external access.

You can allow function calls by defining the following in a resource-based policy.

  • User of the specified Samsung Cloud Platform account
  • Specified source IP address range or CIDR block

A source policy is defined as a JSON policy document attached to the API, which controls whether the specified security principal (typically an IAM role or group) can call the API.

Categorydescriptionexample
PrincipalSpecify the caller of the functionSpecific object storage bucket, API Gateway, other Samsung Cloud Platform accounts, etc.
Task(Action)Define the allowed functionsMostly scf:InvokeFunction
Condition(Condition)Restrict to allow only in specific situationsAllow only requests originating from a bucket with a specific SRN.
Table. Entity that controls API call execution
Reference
  • Cloud Functions’ resource-based policies leverage the rules of IAM’s resource-based policies.
  • For instructions on creating or modifying policies using JSON, refer to the JSON Mode Utilization Guide.

Resource-based policy usage scenario

The primary use cases for resource-based policies are as follows.

Resource-based policy scenario

The resource-based policy scenarios used when a Cloud Functions function runs are as follows.

CategorydescriptionReference example
Function URL - Authentication Type NoneIt is required when generating a function URL for invocation.
  • If there is no resource-based policy, authorization fails, making it impossible to invoke the function using a public URL.
Function URL (Auth Type None) Example
Function URL - Authentication Type IAM
  • Resource-based policy is registered or credential permission is required.
  • If the resource belongs to the same account: you can invoke it if you have a resource-based policy or credential permission (InvokeFunctionUrl).
  • If the resource belongs to a different account: you must have both a resource-based policy and credential permission (InvokeFunction) to invoke it.
Function URL (authentication type IAM) example
API Gateway triggerIt is required when API Gateway calls Lambda to handle external API requests.
  • Resource-based policy registration is required.
  • If a resource-based policy is not present, authorization fails, making it impossible to invoke the function via API Gateway.
API Gateway Trigger Example
PrivateLink connectionYou can connect a PrivateLink Service to define the function for private access from another VPC.
  • It must be registered with a resource-based policy or require credential permissions.
  • When the resource is in the same account: you can invoke it if you have a resource-based policy or credential permission (InvokeFunction).
  • When the resource is in a different account: you must have both a resource-based policy and credential permission (InvokeFunction) to invoke it.
PrivateLink connection example
Table. Resource-based policy scenario

User addition usage scenario

Although it is not automatically registered as a resource-based policy for Cloud Functions, users can add and use it as needed. The scenarios that users can add and utilize are as follows.

  • Cross-Account Access
    • If an IAM user in account A wants to invoke a Lambda in account B, register account A in the function policy of account B.
  • Hybrid Access Control
    • It can be configured so that access is allowed only when both conditions are met—a specific user and a specific IP range—rather than merely restricting by account or IP alone.

Resource-based policy management for Cloud Functions

To view and configure resource-based policies for Cloud Functions, follow these steps.

  1. Click the All Services > Compute > Cloud Functions menu. Navigate to the Service Home page of Cloud Functions.
  2. On the Service Home page, click the Function menu. You will be taken to the Function list page.
  3. On the Function List page, click the resource for which you want to set a policy. You will be taken to the Function Details page.
  4. Click the Configuration tab on the Function Details page.
  5. Click the Edit button of the Resource-based policy permission item. The Resource policy edit popup window opens.
  6. In the Resource Policy edit popup, after selecting the Policy Template, write the policy.
  7. When the writing is complete, click the Confirm button.
    • Click the Delete button to delete the registered policy.

Example of resource-based policy

Users can define additional resource-based policies as needed or modify existing policies for use.

Reference
  • For some features, a resource‑based policy (or credential) must be registered to use them in Cloud Functions.
  • In the resource-based policy examples described in this guide, Cloud Functions automatically registers the example resource-based policies when each feature is enabled or linked.

Function URL - Authentication Type None

Principal is /* a policy that allows public calls.

Policy Template

Color mode
{
    "Statement": [
        {
        	"Action": ["scf:InvokeFunctionUrl"],
        	"Condition": {
            	"StringEquals": {
                	"scf:CloudFunctionAuthType": ["NONE"]
            	}
        	},
        	"Effect": "Allow"
        	"Principal": "*"
            "Resource": ["{{CloudFunctionSrn}}"],
            "Sid": "InvokeFunctionURLAllowPublicAccess"
        }
    ],
    "Version": "2024-07-01"
}
{
    "Statement": [
        {
        	"Action": ["scf:InvokeFunctionUrl"],
        	"Condition": {
            	"StringEquals": {
                	"scf:CloudFunctionAuthType": ["NONE"]
            	}
        	},
        	"Effect": "Allow"
        	"Principal": "*"
            "Resource": ["{{CloudFunctionSrn}}"],
            "Sid": "InvokeFunctionURLAllowPublicAccess"
        }
    ],
    "Version": "2024-07-01"
}
Function URL – Authentication Type None Policy Template Example

Policy example

Color mode
{
    "Statement": [
        {
            "Action": ["scf:InvokeFunctionUrl"],
            "Condition": {
                "StringEquals": {
                    "scf:CloudFunctionAuthType": ["NONE"]
                }
            },
            "Effect": "Allow"
            "Principal": "*"
            "Resource": ["srn:e::accountID:kr-west1::scf:cloud-function/functionsID"],
            "Sid": "InvokeFunctionURLAllowPublicAccess"
        }
    ],
    "Version": "2024-07-01"
}
{
    "Statement": [
        {
            "Action": ["scf:InvokeFunctionUrl"],
            "Condition": {
                "StringEquals": {
                    "scf:CloudFunctionAuthType": ["NONE"]
                }
            },
            "Effect": "Allow"
            "Principal": "*"
            "Resource": ["srn:e::accountID:kr-west1::scf:cloud-function/functionsID"],
            "Sid": "InvokeFunctionURLAllowPublicAccess"
        }
    ],
    "Version": "2024-07-01"
}
Function URL - Authentication Type None Policy Example

Function URL - Authentication Type IAM

This policy permits a specific user to invoke a public URL.

Policy Template

Color mode
{
    "Statement": [
        {
            "Action": ["scf:InvokeFunctionUrl"],
            "Condition": {
                "StringEquals": {
                    "scf:CloudFunctionAuthType": ["SCP_IAM"]
                }
            },
            "Effect": "Allow"
            "Principal": {
                "scp": ["srn:{{Environment}}::{{AccountID}}:::iam:user/{{UserId}}"]
            },
            "Resource": ["{{CloudFunctionSrn}}"],
            "Sid": "Statement1"
        }
    ],
    "Version": "2024-07-01"
}
{
    "Statement": [
        {
            "Action": ["scf:InvokeFunctionUrl"],
            "Condition": {
                "StringEquals": {
                    "scf:CloudFunctionAuthType": ["SCP_IAM"]
                }
            },
            "Effect": "Allow"
            "Principal": {
                "scp": ["srn:{{Environment}}::{{AccountID}}:::iam:user/{{UserId}}"]
            },
            "Resource": ["{{CloudFunctionSrn}}"],
            "Sid": "Statement1"
        }
    ],
    "Version": "2024-07-01"
}
Function URL - Authentication Type IAM Policy Template Example

Policy Example

Color mode
{
    "Statement": [
        {
            "Action": ["scf:InvokeFunctionUrl"],
            "Condition": {
                "StringEquals": {
                    "scf:CloudFunctionAuthType": ["SCP_IAM"]
                }
            },
            "Effect": "Allow"
            "Principal": "*",
            "Resource":  ["srn:e::accountID:kr-west1::scf:cloud-function/functionsID"],
            "Sid": "accountID-iam-invokefunctionurl"
        }
    ],
    "Version": "2024-07-01"
}
{
    "Statement": [
        {
            "Action": ["scf:InvokeFunctionUrl"],
            "Condition": {
                "StringEquals": {
                    "scf:CloudFunctionAuthType": ["SCP_IAM"]
                }
            },
            "Effect": "Allow"
            "Principal": "*",
            "Resource":  ["srn:e::accountID:kr-west1::scf:cloud-function/functionsID"],
            "Sid": "accountID-iam-invokefunctionurl"
        }
    ],
    "Version": "2024-07-01"
}
Function URL - Authentication Type IAM Policy Example

API Gateway trigger

Principal is a policy that permits public calls with a * principal.

Policy Template

Color mode
{
    "Statement": [
        {
            "Action": ["scf:InvokeFunction"],
            "Condition": {
                "SrnLike": {
                    "scp:RequestAttribute/body['x-scf-request-obj-srn']": ["{{ApiGatewayMethodSrn}}"]
                }
            },
            "Effect": "Allow",
            "Principal": {
                "Service": ["apigateway.samsungsdscloud.com"]
            },
            "Resource": ["{{CloudFunctionSrn}}"],
            "Sid": "Statement1"
        }
    ],
    "Version": "2024-07-01"
}
{
    "Statement": [
        {
            "Action": ["scf:InvokeFunction"],
            "Condition": {
                "SrnLike": {
                    "scp:RequestAttribute/body['x-scf-request-obj-srn']": ["{{ApiGatewayMethodSrn}}"]
                }
            },
            "Effect": "Allow",
            "Principal": {
                "Service": ["apigateway.samsungsdscloud.com"]
            },
            "Resource": ["{{CloudFunctionSrn}}"],
            "Sid": "Statement1"
        }
    ],
    "Version": "2024-07-01"
}
API Gateway trigger policy template example

Policy Example

Color mode
{
    "Statement": [
        {
            "Action": [
                scf:InvokeFunction
            ],
            "Condition": {
                "SrnLike": {
                    "scp:RequestAttribute/body['x-scf-request-obj-srn']": [
                        srn:e::accountID:kr-west1::apigateway:method/MethodID/*/GET/test
                    ]
                }
            },
            "Effect": "Allow"
            "Principal": {
                "Service": [
                    "apigateway.samsungsdscloud.com"
                ]
            },
            "Resource": [
                srn:e::accountID:kr-west1::scf:cloud-function/functionID
            ],
            "Sid": "999e9a9999de4d4683c9e10c74ee999z"
        }
    ],
    "Version": "2024-07-01"
}
{
    "Statement": [
        {
            "Action": [
                scf:InvokeFunction
            ],
            "Condition": {
                "SrnLike": {
                    "scp:RequestAttribute/body['x-scf-request-obj-srn']": [
                        srn:e::accountID:kr-west1::apigateway:method/MethodID/*/GET/test
                    ]
                }
            },
            "Effect": "Allow"
            "Principal": {
                "Service": [
                    "apigateway.samsungsdscloud.com"
                ]
            },
            "Resource": [
                srn:e::accountID:kr-west1::scf:cloud-function/functionID
            ],
            "Sid": "999e9a9999de4d4683c9e10c74ee999z"
        }
    ],
    "Version": "2024-07-01"
}
API Gateway trigger policy example

PrivateLink connection

This is a policy that allows function calls through a Privatelink Endpoint for specific users.

Policy Template

Color mode
{
    "Statement": [
        {
            "Action": ["scf:InvokeFunction"],
            "Condition": {
                "StringEquals": {
           			"scf:CloudFunctionPrivatelinkServiceAuthType": ["SCP_IAM"]
                }
            },
            "Effect": "Allow"
            "Principal": {
                "scp": ["srn:{{Environment}}::{{AccountID}}:::iam:user/{{UserId}}"]
            },
            "Resource": ["{{CloudFunctionSrn}}"],
            "Sid": "Statement1"
        }
    ],
    "Version": "2024-07-01"
}
{
    "Statement": [
        {
            "Action": ["scf:InvokeFunction"],
            "Condition": {
                "StringEquals": {
           			"scf:CloudFunctionPrivatelinkServiceAuthType": ["SCP_IAM"]
                }
            },
            "Effect": "Allow"
            "Principal": {
                "scp": ["srn:{{Environment}}::{{AccountID}}:::iam:user/{{UserId}}"]
            },
            "Resource": ["{{CloudFunctionSrn}}"],
            "Sid": "Statement1"
        }
    ],
    "Version": "2024-07-01"
}
Example of a PrivateLink connection policy template

Policy Example

Color mode
{
    "Statement": [
        {
            "Action": [
                scf:InvokeFunction
            ],
            "Condition": {
                "StringEquals": {
                    "scf:CloudFunctionAuthType": [
                        SCP_IAM
                    ]
                }
            },
            "Effect": "Allow",
            "Principal": {
                "scp": [
                    srn:e::accountID:::iam:user/userID
                ]
            },
            "Resource": [
                srn:e::accountID:kr-west1::scf:cloud-function/functionID
            ],
"Sid": "rrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr-privatelink-invokefunction"
        }
    ],
    "Version": "2024-07-01"
}
{
    "Statement": [
        {
            "Action": [
                scf:InvokeFunction
            ],
            "Condition": {
                "StringEquals": {
                    "scf:CloudFunctionAuthType": [
                        SCP_IAM
                    ]
                }
            },
            "Effect": "Allow",
            "Principal": {
                "scp": [
                    srn:e::accountID:::iam:user/userID
                ]
            },
            "Resource": [
                srn:e::accountID:kr-west1::scf:cloud-function/functionID
            ],
"Sid": "rrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrrr-privatelink-invokefunction"
        }
    ],
    "Version": "2024-07-01"
}
Example of PrivateLink connection policy

6.3 - API Reference

API Reference

6.4 - CLI Reference

CLI Reference

6.5 - Release Note

Cloud Functions

2026.05.21
CHANGED Added information on the termination of support for each Runtime version and discontinued AIOS integration
  • Technical support end-of-life information (EoTS) for the Runtime version is provided.
    • When technical support ends, security patches and updates will no longer be applied to that Runtime version.
    • Already created functions can be used continuously without any call time limit. * However, we do not guarantee bugs, errors, defects, or vulnerabilities that arise in Runtime versions for which support has ended.
    • End of technical support scheduled The Runtime versions are as follows (For the full list of supported Runtime versions, see Rutime)
      • GO: 1.21, 1.23
      • Node.js: 18, 20
      • PHP 8.1
      • Python: 3.9
    • Including runtime of additionally excluded functions
  • The AIOS service integration has been stopped.
    • Since the AIOS service has been discontinued, the AIOS integration service has been removed from Cloud Functions.
2026.03.19
FEATURE Add resource-based policy feature
  • You can set a resource-based policy on the function.
    • Resource-based policies are policies applied directly to the function that permit external access.
    • You can use resource-based policies to allow or deny actions on specific resources to specific principals.
2025.12.16
FEATURE AIOS, PrivateLink service integration
  • You can use functions in conjunction with the AIOS service.
    • You can integrate Cloud Functions with AIOS to utilize LLMs.
  • You can use the function in conjunction with the PrivateLink service.
    • Through a Private connection (PrivateLink), you can establish internal connections between VPCs and between a VPC and a service on the Samsung Cloud Platform without traversing the Internet.
2025.10.23
FEATURE Add Java Runtime executable file upload feature
  • The feature to upload Java Runtime executable files has been added.
    • You can import and configure a Java Runtime executable archive file (.jar/.zip) in Object Storage.
2025.07.01
NEW Cloud Functions official version launch
  • We have officially launched the Cloud Functions service.
    • It is a serverless computing‑based FaaS (Function as a Service) that lets you easily run function‑style applications without server provisioning.

7 - Virtual Server DR

When the system is disrupted due to various disaster situations and risk factors, you can replicate the Block Storage attached to a Virtual Server in another region to restore normal operation quickly.

7.1 - Overview

Service Overview

Virtual Server DR is a service that quickly restores the system by replicating the Virtual Server and its attached Block Storage to a region different from the one currently in use. Even if the system is disrupted due to various disaster situations or unexpected events, you can quickly restore normal operation by using Virtual Server DR.

Notice
  • The Virtual Server DR service can be configured using partner solutions sold on the Samsung Cloud Platform Marketplace.
  • For detailed information on using the Marketplace, please refer to Marketplace.
Caution
  • When you purchase and use a service sold on the Marketplace, a delegated tax invoice is issued in accordance with a separate agreement with the Marketplace software supplier.
  • If you request a partner solution product for Virtual Server DR from the Marketplace, the request information will be emailed to the responsible person. Coordinate product details and schedule with the person in charge. Software installation and charges will be billed based on the confirmed date.
  • The services sold in the Marketplace of Samsung Cloud Platform are offered by individual sellers, and Samsung SDS serves as an e‑commerce intermediary, not a party to the sales. Accordingly, Samsung SDS does not guarantee or assume responsibility for the service information or transactions conducted by individual sellers.

Features

  • Easy DR Environment Setup: You can easily configure a Virtual Server for DR setup through partner solutions in the Marketplace of Samsung Cloud Platform.
  • Various Environment Configurations: Using partner solutions, you can configure various environments such as physical to virtual (P2V) and virtual to virtual (V2V) environments, and it supports multiple operating systems (Windows, Linux).

Service Architecture Diagram

Diagram
Figure. Virtual Server DR Diagram

Provided features

For the main features, please refer to the Product Catalog Details page of the partner solution sold on the Samsung Cloud Platform Marketplace.

Note
For detailed information on using the Marketplace, see Marketplace.

Preliminary Service

This is a list of services that must be pre-configured before creating the service. Please refer to the guide provided for each service and prepare accordingly.

Service CategoryserviceDetailed description
NetworkingVPCA service that provides an isolated virtual network in a cloud environment
NetworkingSecurity GroupVirtual firewall that controls server traffic
ComputeVirtual ServerVirtual server optimized for cloud computing
Table. Virtual Server DR Preliminary Service

7.2 - Release Note

Virtual Server DR

2025.07.01
NEW Official launch of Virtual Server DR service
  • We have officially launched the Virtual Server DR service.
    • When the system is interrupted by various disaster scenarios and risk factors, it can be restored to normal operation in a short time.

8 - Block Storage

8.1 - Overview

Service Overview

Block Storage is a high-performance storage that stores data in block units arranged in a fixed size and order.
Suitable for large-scale, high-performance requirements such as databases and mail servers, and users can directly allocate volumes to the server.

Features

  • Large Volume Provisioning: OS configuration volumes are created with at least the minimum size per image and can be expanded up to 12TB, and data storage volumes outside the OS can be created and expanded from a minimum of 8GB to a maximum of 12TB. * Capacity expansion is carried out reliably while online.
  • Full SSD-based high performance: Provides high durability and availability based on redundant Controllers and Disk Array RAID. * Full SSD disks are provided by default, making them suitable for high‑speed data processing tasks such as database workloads.
  • Snapshot Backup: Image snapshot functionality allows recovery of changed and deleted data. * The user selects the snapshot created at the desired recovery point from the list and performs the recovery.

Service architecture diagram

Diagram
Figure. Block Storage diagram

Provided features

Block Storage provides the following features.

  • Volume Name: Users can set or edit the name for each volume.
  • Capacity: Volumes can be created with capacities ranging from a minimum of 8 GB up to a maximum of 12 TB, and can be expanded while in use. * The default OS volume can be created with a size no less than the minimum required for each image.
  • Connection Server: You can select a Virtual Server to connect or disconnect.
    • Multi-Server Connection (Multi Attach): Connects to two or more servers, with no limit on the number of servers per volume, and a Virtual Server can connect up to 26 volumes
  • Encryption: All volumes of Block Storage have AES-256 algorithm encryption applied by default, and when a volume is of HDD/SSD_KMS disk type, additional in‑transit encryption is provided for the Block Storage segment connected to the instance.
  • snapshot: Through the image snapshot feature, recovery of modified and deleted data is possible. * The user selects a snapshot created at the desired recovery point from the list and restores it.
  • Volume Transfer: Using the volume transfer feature, you can transfer a volume to another Account.
  • Monitoring: IOPS, Latency, Throughput, etc. monitoring information can be viewed through the ServiceWatch service.

Components

You can create a volume by entering the capacity and selecting the disk type according to the user’s service scale and performance requirements. When using the snapshot feature, you can restore data to the desired point in time.

Volume

A volume (Volume) is the basic creation unit of the Block Storage service and is used as data storage space. The user selects a name, size, and disk type to create a volume, then attaches it to a Virtual Server for use.
The volume name creation rules are as follows.

Reference
Use English letters, numbers, spaces, and special characters(-, _) within 255 characters.

Snapshot

A snapshot (Snapshot) is an image backup of a volume at a specific point in time. The user can view the snapshot name and creation time in the snapshot list, select the snapshot they wish to restore, and recover data that was modified or deleted using that snapshot.
The following are considerations when using snapshots.

Reference
  • The snapshot creation time is based on Asia/Seoul (GMT +09:00).
  • Select the snapshot recovery button to restore the Block Storage volume to the latest snapshot.
  • If you select a specific snapshot from the snapshot list, you can recover by creating a new volume based on that snapshot.
  • Snapshots are billed based on the size of the original Block Storage, so please delete any unnecessary snapshots.

Preceding Service

This is a list of services that must be pre-configured before creating the service. Refer to the guide provided for each service for detailed information and prepare in advance.

Service CategoryserviceDetailed description
ComputeVirtual ServerVirtual server optimized for cloud computing
Table. Block Storage pre-service

8.1.1 - Monitoring Metrics

Cloud Monitoring service termination notice

According to Samsung Cloud Platform’s policy, the Cloud Monitoring service is scheduled to be discontinued in September 2026.
Accordingly, after the September 2026 release, resource monitoring of the Samsung Cloud Platform via Cloud Monitoring will no longer be possible.

With the new alternative service, you can continuously perform resource monitoring by leveraging ServiceWatch released in October 2025.
ServiceWatch provides more modern and powerful features, replacing Cloud Monitoring to deliver a seamless monitoring environment.

Block Storage planned to be integrated with ServiceWatch from the July 2026 release onward is.
Detailed information about ServiceWatch is available in the ServiceWatch Overview.

Block Storage Monitoring Metrics

The table below shows the monitoring metrics for Block Storage that can be viewed in Cloud Monitoring. For detailed usage of Cloud Monitoring, refer to the Cloud Monitoring guide.

Performance Item NameExplanationunit
Volume TotalTotal byte countbytes
IOPS [Read]iops(read)iops
IOPS [Write]iops(write)iops
Latency Time [Read]Read latencyusec
Latency Time [write]Latency (write)usec
Throughput [Read]Throughput (read)bytes/s
Throughput [Write]Throughput (write)bytes/s
Table. Block Storage Monitoring Metrics

8.1.2 - ServiceWatch metric

Block Storage VM sends metrics to ServiceWatch. The metrics provided by basic monitoring are data collected at 1‑minute intervals.

Reference
For how to view metrics in ServiceWatch, refer to the ServiceWatch guide.

basic metrics

The following are the basic metrics for the Block Storage VM namespace.

The indicators whose names are displayed in bold below are the key indicators selected from the basic metrics provided by Block Storage VM. The key metrics are used to configure service dashboards that are automatically built for each service in ServiceWatch.

Each metric offers guidance in the user guide on which statistical value is meaningful when viewing it, and the service dashboard allows you to view key metrics using their primary statistical values.

Indicator nameDetailed descriptionunitmeaningful statistics
Volume TotalBlock Storage sizeGigabytes-
IOPS [Read]Block Storage volume IOPS (read)-
  • Maximum
  • Average
  • Minimum
IOPS [Write]Block Storage volume IOPS (write)-
  • Maximum
  • Average
  • Minimum
Latency Time [Read]Block Storage volume latency (read)Milliseconds
  • Maximum
  • Average
  • Minimum
Latency Time [write]Block Storage volume latency (write)Millisecond
  • Maximum
  • Average
  • Minimum
Throughput [Read]Block Storage volume throughput (read)Megabytes/Second
  • Maximum
  • Average
  • Minimum
Throughput [Write]Block Storage volume throughput (write)Megabytes/Second
  • Maximum
  • Average
  • Minimum
Table. Block Storage VM basic metrics

8.2 - How-to guides

Users can create the service by entering the required Block Storage information and selecting detailed options through the Samsung Cloud Platform Console.

Create Block Storage

You can create and use the Block Storage service from the Samsung Cloud Platform Console. To create Block Storage, follow these steps.

  1. All Services > Compute > Virtual Server menu, click it. 1. Navigate to the Service Home page of the Virtual Server.

  2. Click the Block Storage menu. 2. Go to the Block Storage List page.

  3. Block Storage on the page, click the Create Service button. 3. Go to the Create Block Storage page.

  4. On the Block Storage Creation page, enter the information required to create the service and select detailed options.

    구분
    Required status
    Detailed description
    Volume nameRequiredVolume name
    • English letters, numbers, spaces, and special characters (-, _) up to 255 characters
    Disk typeRequiredSelect Disk Type
    • SSD_Provisioned: SSD volume with configurable IOPS and throughput
    • SSD/HDD: Standard SSD/HDD volume
    • SSD/HDD_KMS: Additional encrypted volume
    • SSD/HDD_MultiAttach: Volume that can be attached to more than one server
    • Cannot be modified after service creation
    • When creating the service via snapshot restore volume creation, it is set identical to the original and cannot be modified
    capacitySelectionCapacity setting
    • Can be created within 8~12,228 GB
    • Enter the number of units provided in 8 GB increments
    • When creating a service via snapshot recovery volume creation, input a capacity that is equal to or larger than the original
    Max IOPSRequiredEnter the maximum IOPS value between 5,000 and 20,000
    • disk type can be set only when it is SSD_Provisioned
    Max ThroughputRequiredEnter the maximum Throughput value between 250 and 1,000
    • disk type can be set only when it is SSD_Provisioned
    Availability ZoneRequiredSelect the Availability Zone to create resources within the region
    SnapshotSelectionSelect the snapshot to use when creating a volume from a snapshot
    • Use item after checking, snapshot selection is possible
    • When creating a service via snapshot recovery volume creation, provide the recovery snapshot name
    • If not selected, an empty volume is created
    Table. Block Storage service information input items

  5. Summary Verify the detailed information and estimated charges generated in the panel, then click the Create button.

    • When creation is complete, verify the created resource on the Block Storage List page.
Reference
  • You can only attach a volume to a Virtual Server in the same Availability Zone.
  • The performance metrics (IOPS, Throughput) of the configured storage are based on maximum values and do not guarantee consistent values.
  • All volumes of Block Storage are encrypted by default using the AES-256 algorithm.
  • Virtual Servers based on Windows cannot use MultiAttach disks. * Use a separate replication method or solution.
  • If the volume is of the HDD/SSD_KMS disk type, additional in‑transit encryption is provided for the instance and the Block Storage segment attached to the instance.
Caution
HDD/SSD_KMS Using this disk type may result in approximately 60% performance degradation.

Check Block Storage detailed information

You can view and edit the full resource list and detailed information of the Block Storage service. Block Storage Details page consists of Detailed Information, Snapshot List, Tags, Operation History tabs.

To view detailed information about the Block Storage service, follow these steps.

  1. All Services > Compute > Virtual Server menu, click it. 1. Navigate to the Service Home page of the Virtual Server.
  2. Click the Block Storage menu. 2. Navigate to the Block Storage List page.
  3. Block Storage List page, click the resource to view detailed information. 3. Block Storage Details Navigate to the page.
    • Block Storage Details page displays status information and additional feature information, and consists of Details, Snapshot List, Tags, Activity Log tabs.
      CategoryDetailed description
      Volume statusVolume status
      • Creating: Creating
      • Downloading: Creating (applying OS image)
      • Available: Created, server connection available
      • Reserved: Server connection pending
      • Attaching: Connecting to server
      • Detaching: Detaching from server
      • In Use: Server connection established
      • Deleting: Terminating service
      • Awaiting Transfer: Waiting for volume transfer
      • Extending: Extending capacity
      • Error Extending: Abnormal state during capacity extension
      • Backing Up: Backing up volume
      • Restoring Backup: Restoring volume backup
      • Error Backing Up: Abnormal state during volume backup
      • Error Restoring: Error restoring volume backup
      • Error Deleting: Error deleting
      • Error Managing: Abnormal state
      • Error: Abnormal state
      • Maintenance: Temporary maintenance
      • Reverting: Reverting snapshot
      Volume TransferMove the volume to another Account
      Create snapshotImmediately create a snapshot at the time of creation
      Snapshot recoveryRecover the volume using the latest snapshot in Available state
      Delete Block StorageButton to delete the service
      Table. Status information and additional features

Detailed Information

Block Storage List page lets you view detailed information of the selected resource and modify it if needed.

CategoryDetailed description
serviceservice group
Resource TypeResource Type
SRNUnique resource ID in Samsung Cloud Platform
  • In the Block Storage service, it refers to the volume SRN
Resource NameResource name
  • In the Block Storage service, it refers to the volume name
Resource IDService’s unique resource ID
ConstructorUser who created the service
Creation date and timeService creation date and time
ModifierUser who modified the service
Modification date and timeService modification date and time
Volume nameVolume name
  • If you need to edit the volume name, click the Edit button
Volume IDVolume unique ID
Availability ZoneAvailability Zone information
Disk typeDisk type
typeClassification by volume creation method and usage
capacityVolume capacity
  • Can be expanded up to 12,288 GB in increments of multiples of 8
  • Capacity reduction is not possible
  • If capacity expansion is required, click the Edit button
Max IOPSRequired
Max ThroughputRequired
Connection serverConnected Virtual Server
  • Server ID: Unique server ID
  • Server Name: Server name
  • Status: Server status
  • Connection Info: Volume connection information for server connection
  • When adding a Virtual Server connection, click the Add button
  • When removing a Virtual Server connection, click the Disconnect button
Table. Block Storage detailed information items

Snapshot List

Block Storage List page allows you to view the snapshot of the selected resource.

CategoryDetailed description
Snapshot nameSnapshot name
ExplanationSnapshot description
Volume capacityCapacity of the original Block Storage volume for the snapshot target
  • If the original volume capacity and the snapshot volume capacity differ, only recovery using the recovery volume creation method is possible
Creation date and timeSnapshot creation time
StatusSnapshot status
  • Available: Creation completed, recoverable
  • Creating: Creating
  • Error: Abnormal state
  • Deleting: Deleting
  • Error Deleting: Abnormal state while deleting
  • Restoring: Restoring
  • Backing Up: Backing Up
Additional features > MoreSnapshot management button
DeleteSelect the snapshots to delete from the snapshot list and delete them all at once.
Table. Snapshot list tab detailed information items
Caution
  • Snapshots can affect volume capacity management. * Delete unnecessary snapshots after use.
  • Snapshot recovery can be performed while the server is not connected.
Reference
  • The snapshot creation time is based on Asia/Seoul (GMT +09:00).
  • When you click the snapshot recovery button, the volume is restored to the latest snapshot in the Available state.
  • When you select to create a recovery volume on the snapshot list page, a new volume based on the snapshot is created without modifying the existing volume.

Tag

On the Block Storage list page, you can view the tag information of the selected resource and add, modify, or delete it.

CategoryDetailed description
Tag listTag list
  • You can view the Key and Value information of the tag
  • Up to 50 tags can be added per resource
  • When entering tags, search the existing list of created Keys and Values and select
Table. Block Storage tag tab item

Operation History

Block Storage List page allows you to view the operation history of the selected resource.

CategoryDetailed description
Task History ListResource Change History
  • Operation Time, Resource Type, Resource ID, Resource Name, Operation Details, Event Topic, Operation Result, Operator Information
  • Click the Advanced Search button to perform a detailed search
Table. Work History tab detailed information items

Block Storage Resource Management

If you need to modify the settings of a created Block Storage or add or remove a connected server, you can perform the operation on the Block Storage Details page.

Edit volume name

You can edit the volume name. To modify the volume name, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. 1. Go to the Service Home page of the Virtual Server.
  2. Click the Block Storage menu. 2. Go to the Block Storage List page.
  3. On the Block Storage List page, click the resource whose volume name you want to edit. 3. Go to the Block Storage Details page.
  4. Click the Edit button of the Volume Name. 4. Edit Volume Name The popup window opens.
  5. Enter the volume name and click the Confirm button.
Reference
Use English letters, numbers, spaces, and special characters (-, _) within 255 characters.

Expand Capacity

You can expand the volume capacity. To increase capacity, follow the steps below.

  1. All Services > Compute > Virtual Server menu, click it. 1. Go to the Service Home page of the Virtual Server.
  2. Click the Block Storage menu. 2. Go to the Block Storage List page.
  3. On the Block Storage List page, click the resource to increase capacity. 3. Block Storage Details page.
  4. Click the Edit button of Capacity. 4. Capacity Modification popup window opens.
  5. Enter the capacity and click the Confirm button.
Caution
용량 축소는 제공하지 않습니다. 용량 증설 후에는 증설 전의 스냅샷으로 복구되지 않습니다. 용량 증설 이전에 생성한 스냅샷으로는 신규 볼륨 생성 방식의 복구만 가능합니다.
Reference
8~12,228GB 이내에서 기존 용량보다 큰 용량으로 증설 가능합니다. 8GB 단위로 제공되는 Unit 개수를 입력하세요.

연결 서버 수정하기 {#update-1} 서버를 연결하거나 연결 해제할 수 있습니다. 연결 서버를 수정 하려면 다음 절차를 따르세요. 모든 서비스 > Compute > Virtual Server 메뉴를 클릭하세요. Virtual Server의 Service Home 페이지로 이동합니다. Block Storage 메뉴를 클릭하세요. Block Storage 목록 페이지로 이동합니다. Block Storage 목록 페이지에서 연결 서버를 수정할 자원을 클릭하세요. 3. Block Storage Details Navigate to the page. 4. When adding a Virtual Server connection, click the Add button in the Connection Server item. 4. Add Connection Server The popup window opens. 5. After selecting the Virtual Server you want to connect to, click the Confirm button. 6. If you disconnect the Virtual Server, click the Disconnect button in the Connection Server item. * Be sure to perform the disconnect operation (Umount, Disk Offline) on the server before proceeding with the disconnection.

Caution
Log in to the server, be sure to perform the disconnect operations (Umount, Disk Offline), and then detach the connected server. Releasing without OS actions may cause a status error (Hang) on the connected server. For detailed information about server disconnection, see 서버 연결 해제하기.
Reference
  • You can connect a Virtual Server created in the same location as the Block Storage.
  • Virtual servers that use a Partition with a Server Group policy cannot be connected.
  • For HDD/SSD_MultiAttach disk types, they can be attached to two or more Virtual Servers, and there is no limit on the number of connections.
    • Virtual Servers based on Windows cannot use MultiAttach disks and must use a separate replication method or solution.
  • Virtual Server can connect up to 26 volumes, including the OS default.
  • The OS default volume cannot be modified on the connected server, nor can its service be deleted.
  • When adding a connected server, it can be used after performing the connection tasks (Mount, Disk Online) on the server. * For detailed information about server connections, refer to 서버 연결하기.

Delete Block Storage

You can reduce operating costs by deleting unused Block Storage. Note that deleting a service may cause the running service to stop immediately, so you should proceed with the deletion only after fully considering the impact of service interruption.

Caution
  • Be careful, as the data cannot be recovered after deletion.
  • In the following case, the Block Storage volume cannot be deleted.
    • Connecting to server
    • OS default volume
    • Connect to the Virtual Server’s custom image
    • When the volume status is not Available, Error, Error Extending, Error Restoring, or Error Managing.
  • When selecting and deleting two or more volumes, only the deletable volumes will be deleted.

To delete Block Storage, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. Click the Block Storage menu. 2. Go to the Block Storage List page.
  3. On the Block Storage List page, select the resource to delete, and click the Delete button.
    • After navigating to the detailed page of the resource to be deleted, you can click the Block Storage Delete button to delete it individually.
  4. When the deletion notification pop-up appears, type Delete and click the Confirm button.
  5. After the deletion is complete, verify on the Block Storage List page that the resource has been removed.

8.2.1 - Connect to Server

When using a volume on the server, a connect or disconnect operation is required.
On the Block Storage Details page, add a connected server, then access the server to perform connection operations (Mount, Disk Online). After use, perform disconnection operations (Umount, Disk Offline) and then remove the connected server.

Connect to Server (Mount, Disk Online)

To use the volume added to the connected server, you must log into the server and perform the connection tasks (Mount, Disk Online). Follow the steps below.

Linux operating system

Example configuration for server connection
  • Server OS: LINUX
  • Mount location: /data
  • Volume size: 24 GB
  • File system: ext3, ext4, xfs etc
  • Additional allocated disk: /dev/vdb
  1. Click the All Services > Compute > Virtual Server menu. Navigate to the Service Home page of Virtual Server.
  2. Click the Block Storage menu. Go to the Block Storage List page.
  3. On the Block Storage List page, click the resource to be used on the connected server. Navigate to the Block Storage Details page.
  4. After verifying the server in the Connection Server item, connect.
  5. Refer to the steps below to mount the volume (Mount).
  • Switch to root privileges

    $ sudo -i
    
  • Check Disk

    # lsblk
    NAME    MAJ:MIN  RM   SIZE RO TYPE MOUNTPOINT
    vda       252:0   0    24G  0 disk
    ├─vda1    252:1   0  23.9G  0 part [SWAP]
    └─vda14   252:14  0     4M  0 part /
    └─vda15   252:15  0   106M  0 part /boot/efi
    vdb       252:16  0    24G  0 disk
    
  • Create partition

    # fdisk /dev/vdb
    Command (m for help): n
    
    Partition type:
       p   primary (0 primary, 0 extended, 4 free)
       e   extended
    Select (default p): p
    Partition number (1-4, default 1): 1
    First sector (2048-50331646, default 2048):
    Last sector, +/-sectors or +/-size{K,M,G,T,P} (2048-50331646, default 50331646):
    
    Created a new partition 1 of type 'Linux' and of size 24 GiB.
    
    Command (m for help): w
    The partition table has been altered!
    Calling ioctl() to re-read partition table.
    Syncing disks.
    
  • Partition format setting (example: ext4)

    # lsblk
    NAME    MAJ:MIN  RM   SIZE RO TYPE MOUNTPOINT
    vda       252:0   0    24G  0 disk
    ├─vda1    252:1   0  23.9G  0 part [SWAP]
    └─vda14   252:14  0     4M  0 part /
    └─vda15   252:15  0   106M  0 part /boot/efi
    vdb       252:16  0    24G  0 disk
    └─vdb1    252:17  0    24G  0 part
    
    # mkfs.ext4 /dev/vdb1
    mke2fs 1.46.5 (30-Dec-2021)
    ...
    Writing superblocks and filesystem accounting information: done
    
  • Volume Mount

    # mkdir /data
    
    # mount /dev/vdb1 /data
    
    # lsblk
    NAME    MAJ:MIN  RM   SIZE RO TYPE MOUNTPOINT
    vda       252:0   0    24G  0 disk
    ├─vda1    252:1   0  23.9G  0 part [SWAP]
    └─vda14   252:14  0     4M  0 part /
    └─vda15   252:15  0   106M  0 part /boot/efi
    vdb       252:16  0    24G  0 disk
    └─vdb1    252:17  0    24G  0 part /data
    
    # vi /etc/fstab
    (추가) /dev/vdb1     /data       ext4     defaults   0 0
    

ItemExplanation
cat /etc/fstabFile system information file
  • Used when the server is running
df -hCheck the total disk usage of mounted disks
fdisk -lCheck partition information
  • Physical disks are represented by letters such as a, b, c like /dev/sda, /dev/sdb, /dev/sdc. Disk partitions are represented by numbers such as 1, 2, 3 like /dev/sda1, /dev/sda2, /dev/sda3
Table. Mount command reference
commandExplanation
mCheck the usage of the fdisk command
nCreate a new partition
pVerify the updated partition information
tChange the system ID of the partition
wSave partition information and exit fdisk.
Table. Partition creation command (fdisk) reference

Windows operating system

  1. Click the All Services > Compute > Virtual Server menu. Navigate to the Virtual Server Service Home page.
  2. Click the Block Storage menu. You will be taken to the Block Storage List page.
  3. On the Block Storage List page, click the resource to be used on the connected server. You will be taken to the Block Storage Details page.
  4. Connection Server item, verify the server and then connect.
  5. Refer to the steps below to connect the volume (Disk Online).
  • After right-clicking the Windows Start icon, run Computer Management.

  • In the Computer Management tree structure, select Storage > Disk Management

  • Check Disk

    Mount

  • Disk Online

    Mount

  • Disk initialization

    Mount

    Mount
  • Partition format

    Mount

    Mount
    Mount
  • Check volume

    Mount

Disconnect server (Umount, Disk Offline)

Access the server, perform the disconnect operations (Umount, Disk Offline), and then disconnect the server from the Console.

Follow the next procedure.

Caution
  • If you disconnect the server from the console without performing the disconnect operations (Umount, Disk Offlline) on the server, a server state error (Hang) may occur.
    • Be sure to perform the OS tasks first.
  • For the OS default volume, you cannot modify the connected server or cancel the service.

Linux operating system

  1. Click the All Services > Compute > Virtual Server menu. Navigate to the Service Home page of Virtual Server.
  2. Click the Block Storage menu. Navigate to the Block Storage List page.
  3. On the Block Storage List page, click the resource to detach the attached server. You will be taken to the Block Storage Details page.
  4. After verifying the server in the Connection Server item, connect.
  5. Refer to the steps below to unmount the volume(Umount).
    • Volume Umount
  # umount /dev/vdb1 /data

  # lsblk
  NAME    MAJ:MIN  RM   SIZE RO TYPE MOUNTPOINT
  vda       252:0   0    24G  0 disk
  ├─vda1    252:1   0  23.9G  0 part [SWAP]
  └─vda14   252:14  0     4M  0 part /
  └─vda15   252:15  0   106M  0 part /boot/efi
  vdb       252:16  0    24G  0 disk
  └─vdb1    252:17  0    24G  0 part

  # vi /etc/fstab
  (삭제) /dev/vdb1     /data       ext4     defaults   0 0

Windows operating system

  1. Click the All Services > Compute > Virtual Server menu. Navigate to the Service Home page of Virtual Server.

  2. Click the Block Storage menu. Go to the Block Storage List page.

  3. Block Storage List page, click the resource to detach the connected server. Navigate to the Block Storage Details page.

  4. In the Connection Server field, verify the server before connecting.

  5. Unmount the mounted file system.

  6. Please follow the procedure below to detach the volume (Disk Offline).

    • Right-click the Windows Start icon, then run Computer Management.

    • In the Computer Management tree structure, select Storage > Disk Management

    • Right-click the disk to be removed, then run Offline

      Umount
    • Check disk status

      Umount

8.2.2 - Using Snapshots

You can create a snapshot of the created Block Storage, delete it, or recover using the snapshot. You can perform actions on the Block Storage Details page and the Snapshot List page.

Create Snapshot

You can create a snapshot of the current point in time. To create a snapshot, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. Block Storage Click the menu. 2. Block Storage List navigate to the page.
  3. On the Block Storage List page, click the resource to create a snapshot. 3. Block Storage Details Navigate to the page.
  4. Click the Create Snapshot button. 4. Create Snapshot The popup window opens.
  5. Enter the Snapshot Name and Description, then click the Create button.
  6. When the popup notifying snapshot creation opens, click the Confirm button. 6. Creates a snapshot of the current point in time.
  7. Click the Snapshot List button. 7. Navigate to the Block Storage Snapshot List page.
  8. Check the generated snapshot.
Caution
Snapshots are billed based on the size of the original Block Storage, so please delete any unnecessary snapshots.
Reference
The snapshot creation time is based on Asia/Seoul (GMT +09:00).

Edit Snapshot

You can modify snapshot information. To modify the snapshot name or description, follow these steps.

  1. All Services > Compute > Virtual Server Click the menu. 1. Go to the Service Home page of the Virtual Server.
  2. Click the Block Storage menu. 2. Go to the Block Storage List page.
  3. On the Block Storage List page, click the resource to edit the snapshot information. 3. Block Storage Details Navigate to the page.
  4. Click the Snapshot List button. 4. Navigate to the Block Storage Snapshot List page.
  5. After confirming the snapshot to edit, click the More button.
  6. Edit Click the button. 6. Snapshot Edit The popup window opens.
  7. Enter the Snapshot name or Description, then click the Confirm button.

Restore Snapshot

You can restore a Block Storage volume to the latest snapshot in the Available state. To restore a snapshot, follow these steps.

  1. All Services > Compute > Virtual Server menu, please click. 1. Navigate to the Service Home page of the Virtual Server.
  2. Block Storage Click the menu. 2. Go to the Block Storage List page.
  3. On the Block Storage List page, click the resource you want to restore from a snapshot. 3. Navigate to the Block Storage Details page.
  4. If there is a server added in the Connection Server item, after connecting to the server, perform the disconnect operations (Umount, Disk Offline).
  5. On the Block Storage Details page, click the Disconnect button in the Connected Server item to remove the server. 5. The connection server will be removed.
  6. Click the Snapshot List button. 6. Block Storage Snapshot List Navigate to the page.
  7. Check the latest snapshot in the Available state. 7. The volume will be restored from this snapshot.
  8. Click the Snapshot Recovery button. 8. Snapshot Recovery popup window opens.
  9. After checking the snapshot name and creation timestamp, click the Confirm button.
    • When recovery starts, it becomes Reverting, and when completed, it becomes Available.
  10. Click the Details page button. 10. Go to the Block Storage Detail Page.
  11. Click the Add button of the Connection Server. 11. Reconnect the Virtual Server.
  12. After connecting to the added server, perform the connection tasks (Mount, Disk Online) according to the operating system.
Caution
  • Snapshot recovery can be performed while the server is not connected.
  • If you need to restore from a snapshot that is not the latest, you can recover by creating a recovery volume.
  • Recovery is not possible in the following situations.
    • When a Block Storage volume is not in the Available state
    • If a server is attached to the Block Storage volume
    • If there are no recoverable snapshots.
    • If the latest snapshot changes while creating a recovery.
    • If the latest snapshot is not in the Available state.
    • When the snapshot’s volume size differs from the Block Storage volume size (in case the volume was expanded)

Create a snapshot recovery volume

You can create a volume using a snapshot. To create a snapshot recovery volume, follow these steps.

  1. All Services > Compute > Virtual Server Click the menu. 1. Navigate to the Service Home page of the Virtual Server.

  2. Click the Block Storage menu. 2. Go to the Block Storage List page.

  3. On the Block Storage List page, click the resource to create a snapshot recovery volume. 3. Navigate to the Block Storage Details page.

  4. Click the Snapshot List button. 4. Navigate to the Block Storage Snapshot List page.

  5. After verifying the snapshot name, description and creation date and time, click the More button of the snapshot you wish to restore.

  6. Click Create recovery volume. 6. Create snapshot recovery volume A popup window opens.

  7. Click the Create button. 7. Go to the Create Block Storage page.

  8. On the Block Storage Creation page, enter the information required to create the service and select detailed options.

    • Enter the volume name and capacity. * You can input a capacity that is greater than or equal to the original volume.
    • The disk type is set to match the original and cannot be modified.
      Category
      Required
      상세 설명
      볼륨명필수볼륨 이름
      • 영문, 숫자, 공백과 특수문자(-, _) 사용하여 255자 입력
      디스크 유형필수디스크 유형 선택
      • SSD_Provisioned: IOPS, Throughput 설정이 가능한 SSD 볼륨
      • SSD/HDD: 일반 SSD/HDD 볼륨
      • SSD/HDD_KMS: 추가 암호화 볼륨
      • SSD/HDD_MultiAttach: 2개 이상의 서버 연결이 가능한 볼륨
      • 서비스 생성 후 수정 불가
      • 스냅샷 복구 볼륨 생성을 통해 서비스 생성 시 원본과 동일하게 설정되며 수정이 불가함
      용량선택용량 설정
      • 8~12,228GB 이내로 생성 가능
      • 8GB 단위로 제공되는 Unit 개수 입력
      • 스냅샷 복구 볼륨 생성을 통해 서비스 생성 시 원본보다 크거나 같은 용량으로 입력
      Max IOPS필수Enter the maximum IOPS value between 5,000 and 20,000
      • Disk type can be set only when it is SSD_Provisioned
      Max ThroughputRequiredEnter the maximum Throughput value between 250 and 1,000
      • Can be set only when Disk type is SSD_Provisioned
      Recovery snapshot nameSelectionName of the recovery snapshot used when creating the volume
      • Provide the recovery snapshot name when creating a service via snapshot recovery volume creation
      Table. Block Storage service information entry items
  9. Summary Verify the detailed information and estimated charges generated in the panel, then click the Create button.

    • When creation is complete, check the created resources on the Block Storage List page.

Delete Snapshot

You can select a snapshot and delete it. To delete a snapshot, follow these steps.

  1. All Services > Compute > Virtual Server Click the menu. 1. Go to the Service Home page of the Virtual Server.
  2. Click the Block Storage menu. 2. Go to the Block Storage List page.
  3. On the Block Storage List page, click the resource to delete the snapshot. 3. Block Storage Details Navigate to the page.
  4. Click the Snapshot List button. 4. Navigate to the Block Storage Snapshot List page.
  5. After verifying the snapshot name, description and creation date and time, click the More button of the snapshot you wish to delete.
  6. Click the Delete button. 6. Snapshot List page removes snapshots.

8.2.3 - Transfer volume

You can transfer the volume to another account, and when you do so, the volume is removed from its original location. You can perform volume migration on the Block Storage List or Block Storage Details page.

Previous volume

You can transfer the volume to another Account. To migrate the volume, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. 1. Navigate to the Service Home page of Virtual Server.
  2. Click the Block Storage menu. 2. Block Storage List Navigate to the page.
  3. On the Block Storage List page, after selecting the resource to migrate, click the More > Migrate Volume button at the top left of the list.
    • Or click the Volume Transfer button at the top of the Block Storage Details page of the resource to be transferred.
  4. When the popup notifying volume transfer opens, verify the volume name you want to transfer, then click the Confirm button.
  5. When the previous completion popup opens, click the Confirm button. 5. Volume Transfer ID and Approval Key information will be downloaded as a text file.
  6. The volume changes to Awaiting Transfer status.
Caution
  • You can attach the volume only to a Virtual Server in the same Availability Zone as the migrated volume.
  • Volume migration is possible only when the volume is in Available state. * If it is In Use, disconnect all connected servers.

Cancel previous volume

You can cancel after creating a volume transfer. To cancel the volume transfer, follow these steps.

  1. Click the All Services > Compute > Virtual Server menu. 1. Navigate to the Service Home page of the Virtual Server.
  2. Click the Block Storage menu. 2. Navigate to the Block Storage List page.
  3. On the Block Storage List page, click the resource to cancel the volume migration. 3. Go to the Block Storage Details page.
    • If the volume is in Awaiting Transfer status, it can be canceled.
  4. Click the Cancel previous volume button. 4. Cancel previous volume A popup window opens.
  5. After confirming the volume name for which you want to cancel the volume transfer, click the Confirm button.
  6. The volume changes to Available state.

Receive previous volume

You can receive a volume transfer from another Account. To receive a volume transfer, follow the steps below.

  1. All Services > Compute > Virtual Server menu, click it. 1. Navigate to the Service Home page of the Virtual Server.
  2. Click the Block Storage menu. 2. Go to the Block Storage List page.
  3. On the Block Storage List page, click the More > Receive Volume Transfer button at the top left of the list. 3. Receive previous volume A popup window opens.
  4. Volume Migration Enter the provided Volume Migration ID and Approval Key.
  5. A volume is created on the Block Storage List page.
information
  • It takes time for the changes to take effect.
  • In the account that initiated the volume transfer, the transferred volume is removed.

8.3 - API Reference

API Reference

8.4 - CLI Reference

CLI Reference

8.5 - Release Note

Block Storage

2026.07.16
FEATURE SeviceWatch service integration and other feature additions
  • ServiceWatch service has been integrated.
    • You can view monitoring information such as IOPS, latency, and throughput via the ServiceWatch service.
  • You can select an Availability Zone when creating a resource.
  • A text input feature has been added when deleting resources.
    • To prevent accidentally deleting resources, you must verify the deletion information and enter Delete when deleting to proceed with the deletion operation.
2026.03.19
FEATURE SSD_Provisioned disk type added
  • An SSD volume with configurable IOPS and Throughput has been added.
    • When creating Block Storage, you can select the SSD_Provisioned disk type.
    • You can set the IOPS and Throughput range.
      • IOPS: 5,000 ~ 20,000
      • Throughput: 250 ~ 1,000
    • During the preview period, there is no additional charge; charges will be applied in the future.
2025.07.01
FEATURE Snapshot fee policy change and monitoring integration
  • Snapshots are billed based on the size of the original Block Storage.
  • Integrated with Cloud Monitoring.
    • You can view IOPS, latency, and throughput information in Cloud Monitoring.
2025.02.27
FEATURE Add Block Storage disk type
  • Block Storage feature change
    • HDD disk types have been added, allowing you to select the added type (HDD, HDD_MultiAttach, HDD_KMS) according to the purpose.
  • Samsung Cloud Platform Common Feature Change
    • Account, IAM, Service Home, tags, and other common CX changes have been reflected.
2024.10.01
NEW Block Storage Service Official Version Release
  • Added SSD_KMS disk type.
    • When SSD_KMS is selected, encryption using KMS (Key Management Service) encryption keys is added.
  • We have launched a high-performance storage service suitable for handling large-scale data and database workloads.
2024.07.02
NEW Beta version release
  • We have launched a high-performance storage service suitable for handling large-scale data and database workloads.