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

Return to the regular view of this page.

Overview

1: Server Type
2: Monitoring Metrics
3: ServiceWatch Metrics

Service Overview

Event Streams provides fully managed creation and configuration of the open-source Apache Kafka for large-scale, high-volume message data processing. Samsung Cloud Platform automates the creation and configuration of Apache Kafka through a web-based console, allowing users to configure the main components of Apache Kafka—Broker, Zookeeper, and AKHQ—in either a single or clustered setup.

The Event Streams cluster consists of multiple Broker nodes; you can install between 1 and 10 Brokers, typically deploying three or more. Zookeeper can be installed separately to manage the distributed Brokers, but if not installed separately, it is installed on the Broker nodes. Additionally, we provide AKHQ (Apache Kafka HQ), a tool for managing Kafka, allowing users to perform cluster operation and management through it.

Provided features

Event Streams provides the following features.

Auto Provisioning: You can configure and set up an Apache Kafka cluster via the UI.
Operation Control Management: Provides functionality to control the status of running servers. In addition to starting and stopping the cluster, restarting is possible to apply configuration changes.
AKHQ provision: We provide AKHQ, a tool for managing Kafka, enabling users to manage and monitor clusters.
Add Broker node: If expansion is required to improve cluster performance and stability, you can add a node with the same specifications as the existing Broker nodes.
Parameter management: You can configure and modify parameters related to performance improvement and security.
Monitoring: CPU, memory, performance monitoring information can be accessed via Cloud Monitoring and Servicewatch.

Component

Event Streams provides pre‑validated engine versions and various server types in accordance with its open‑source support policy. Users can select and use them based on the scale of the service they wish to configure.

Engine version

The engine versions supported by Event Streams are as follows.

Technical support can be used until the supplier’s EoTS (End of Technical Service) date, and the EOS date when new creation is stopped is set to six months before the EoTS date.

The EOS and EoTS dates may change according to the supplier’s policy, so please refer to the supplier’s license management policy page for details.

Apache Kafka: https://docs.confluent.io/platform/current/installation/versions-interoperability.html

Provided version	EoS Date	EoTS Date
3.8.0	2026-07 (planned)	2026-12-02
3.9.1	2026-09 (planned)	2027-02-19

Table. Event Streams Supported Engine Versions

Server Type

The server types supported by Event Streams are as follows.

For detailed information about the server types provided by Event Streams, refer to Event Streams Server Types.

Standard ess1v2m4

Category	example	Detailed description
Server type	Standard	Provided server types Standard: Standard configuration (vCPU, Memory) commonly used High Capacity: Large-capacity server specifications of 24 vCores or more
Server specifications	ess1	Provided server specifications ess1, ess2: Standard specifications (vCPU, Memory) commonly used esh2: Large-capacity server specifications Providing servers with 24 vCores or more
Server specifications	v2	Number of vCores v2: 2 virtual cores
Server specifications	m4	Memory capacity m4: 4GB Memory

Table. Event Streams Server Type Components

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 in advance.

Service Category	service	Detailed description
Networking	VPC	A service that provides an isolated virtual network in a cloud environment

Table. Event Streams Preliminary Services

1 - Server Type

Event Streams server type

Event Streams provides server types composed of various combinations such as CPU, Memory, and Network Bandwidth. When creating Event Streams, Apache kafka is installed according to the server type selected for the intended purpose.

The server types supported by Event Streams are as follows.

Standard ess1v2m4

Category	example	Detailed description
Server type	Standard	Provided server type classifications Standard: Configured with commonly used standard specifications (vCPU, Memory) High Capacity: Large-capacity server specifications exceeding Standard
Server specifications	ess1	Provided server type classification and generation ess1: s means standard specification, and 1 indicates the generation esh2: h means high-capacity server specification, and 2 indicates the generation
Server specifications	v2	Number of vCores v2: 2 virtual cores
Server specifications	m4	Memory capacity m4: 4GB Memory

Table. Event Streams server type format

Reference

Check the node’s minimum specifications as shown below and select the server type.

Category	vCPU	Memory
Broker	2 vCore	4 GB
Zookeeper	1 vCore	2 GB

ess1 server type

The ess1 server type of Event Streams is offered with standard specifications (vCPU, Memory) and is suitable for various database workloads.

Intel 3rd‑generation (Ice Lake) Xeon Gold 6342 Processor up to 3.3 GHz
Supports up to 16 vCPUs and 64 GB of memory
Maximum networking speed of 12.5 Gbps

Category	Server type	vCPU	Memory	Network Bandwidth
Standard	ess1v1m2	1 vCore	2 GB	Up to 10 Gbps
Standard	ess1v2m4	2 vCore	4 GB	Up to 10 Gbps
Standard	ess1v2m8	2 vCore	8 GB	Up to 10 Gbps
Standard	ess1v4m8	4 vCore	8 GB	Up to 10 Gbps
Standard	ess1v4m16	4 vCore	16 GB	Up to 10 Gbps
Standard	ess1v8m16	8 vCore	16 GB	Up to 10 Gbps
Standard	ess1v8m32	8 vCore	32 GB	Up to 10 Gbps
Standard	ess1v16m32	16 vCore	32 GB	Up to 12.5 Gbps
Standard	ess1v16m64	16 vCore	64 GB	Up to 12.5 Gbps

Table. Event Streams server type specifications - ess1 server type

ess2 server type

The ess2 server type of Event Streams is offered with standard specifications (vCPU, Memory) and is suitable for various database workloads.

Intel 4th‑generation (Sapphire Rapids) Xeon Gold 6448H Processor up to 3.2 GHz
Supports up to 16 vCPUs and 64 GB of memory
Maximum networking speed of 12.5 Gbps

Category	Server type	CPU vCore	Memory	Network Bandwidth(Gbps)
Standard	ess2v1m2	1 vCore	2 GB	Up to 10 Gbps
Standard	ess2v2m4	2 vCore	4 GB	Up to 10 Gbps
Standard	ess2v2m8	2 vCore	8 GB	Up to 10 Gbps
Standard	ess2v4m8	4 vCore	8 GB	Up to 10 Gbps
Standard	ess2v4m16	4 vCore	16 GB	Up to 10 Gbps
Standard	ess2v8m16	8 vCore	16 GB	Up to 10 Gbps
Standard	ess2v8m32	8 vCore	32 GB	Up to 10 Gbps
Standard	ess2v16m32	16 vCore	32 GB	Up to 12.5 Gbps
Standard	ess2v16m64	16 vCore	64 GB	Up to 12.5 Gbps

Table. Event Streams server type specifications - ess2 server type

esh2 server type

The esh2 server type of Event Streams is offered with high-capacity server specifications and is suitable for database workloads that require large-scale data processing.

Intel 4th‑generation (Sapphire Rapids) Xeon Gold 6448H Processor up to 3.2 GHz
Supports up to 32 vCPUs and 128 GB of memory
Up to 25 Gbps networking speed

Category	Server type	vCPU	Memory	Network Bandwidth
High Capacity	esh2v32m64	32 vCore	64 GB	Up to 25 Gbps
High Capacity	esh2v32m128	32 vCore	128 GB	Maximum 25 Gbps

Table. Event Streams server type specifications - esh2 server type

2 - Monitoring Metrics

Cloud Monitoring Service Termination Notice

According to the policy of Samsung Cloud Platform, the Cloud Monitoring service is scheduled to be terminated.

Accordingly, Event Streams will no longer be able to monitor resources via Cloud Monitoring after the service improvement work in July 2026.
After that, resource monitoring can be continuously performed through the ServiceWatch service, which was released in October 2025.

Resources created after the service improvement work in March 2026 are provided with monitoring based on the ServiceWatch service by default.

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

Event Streams monitoring metrics

The table below shows the performance monitoring metrics for Event Streams that can be viewed through Cloud Monitoring. For detailed Cloud Monitoring usage, see the Cloud Monitoring guide.

For Event Streams server monitoring metrics, refer to the Virtual Server Monitoring Metrics guide.

Performance items	Detailed description	unit
AKHQ State [PID]	AHKQ process PID	PID
Connections [Zookeeper Client]	Number of ZooKeeper connections	cnt
Disk Used	datadir usage	bytes
Failed [Client Fetch Request]	Client fetch request processing failure count	cnt
Failed [Produce Request]	Procucer request processing failure count	cnt
Incomming Messages	Number of messages received by the broker	cnt
Instance State [PID]	kafka process PID	PID
Kibana state [PID]	Kibana process PID	PID
Leader Elections	Number of Leader Election occurrences	cnt
Leader Elections [Unclean]	Number of Unclean Leader Election occurrences	cnt
Log Flushes	Number of log flush occurrences	cnt
Network In Bytes	Total bytes received by the entire Topic	bytes
Network Out Bytes	Total bytes transmitted by the Topic	bytes
Rejected Bytes	Total bytes rejected by the Topic	bytes
Request Queue Length	Request queue size	cnt
Shards	Cluster shard count	cnt
Zookeeper Sessions [Closed]	ZooKeeper closed sessions per second	cnt
Zookeeper Sessions [Expired]	ZooKeeper expired sessions per second	cnt
Zookeeper State [PID]	zookeeper process PID	PID

Table. Event Streams monitoring metrics

3 - ServiceWatch Metrics

Event Streams sends metrics to ServiceWatch. The metrics provided by default monitoring are data collected at 1‑minute intervals.

Note

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

Basic Metrics

The following are the basic metrics for the Event Streams namespace.

Indicators whose names are shown in bold below are the indicators selected as key metrics among the default metrics provided by Event Streams. Key metrics are used to build service dashboards that are automatically created for each service in ServiceWatch.

Each metric provides guidance in the user guide about which statistical value is meaningful when querying that metric, and among the meaningful statistics, the statistical values displayed in bold text are the primary statistics. In the service dashboard, primary metrics can be viewed using the primary statistical values.

OS Basic Metrics

Category	Performance items	unit	Detailed description	meaningful statistics
CPU	CPU Usage	Percent	CPU usage	maximum
Disk	Disk Usage	Percent	Disk usage	Maximum
Disk	Disk Write Bytes	Bytes/Second	Write throughput (bytes/second) on block device	average maximum minimum
Disk	Disk Read Bytes	Bytes/Second	Read capacity (bytes/second) from block device	average maximum minimum
Disk	Disk Write Requests	Count/Second	Number of write requests on block device (requests per second)	average maximum minimum
Disk	Disk Read Requests	Count/Second	Number of read requests on block device (requests per second)	average maximum minimum
Disk	Average Disk I/O Queue Size	None	Average queue length of requests issued to the block device	average maximum minimum
Disk	Disk I/O Utilization	Percent	The proportion of time the block device actually spent handling I/O operations	average maximum minimum
Memory	Memory Usage	Percent	Memory usage	Maximum
Network	Network In Bytes	Bytes/Second	Received data rate (bytes/second) on the network interface	average maximum minimum
Network	Network Out Bytes	Bytes/Second	Transmitted capacity (bytes/second) on the network interface	average maximum minimum
Network	TCP Connections	Count/Second	Current total number of successfully established TCP connections	average maximum minimum
Network	Network In Packets	Count	Number of packets received on the network interface	Average Maximum Minimum
Network	Network Out Packets	Count	Number of packets transmitted on the network interface	average maximum minimum
Network	Network In Dropped	Count	Number of packet drops received on the network interface	average maximum minimum
Network	Network Out Dropped	Count	Number of packet drops transmitted from the network interface	average maximum minimum
Network	Network In Errors	Count	Number of packet errors received on the network interface	average maximum minimum
Network	Network Out Errors	Count	Number of packet errors transmitted on the network interface	average maximum minimum

Table. OS basic metrics

Event Streams basic metrics

Category	Performance items	unit	Detailed description	meaningful statistics
JVM	JVM Threads Current	Count	Number of currently active threads	Total Average Maximum Minimum
JVM	JVM Memory Usage	Bytes	Current amount of memory used by the JVM	Total Average Maximum Minimum
JVM	Time spent in GC	None	JVM time spent on garbage collection (GC seconds per minute)	Total Average Maximum Minimum
JVM	Count in GC	Count/Minute	Number of times the JVM executed the garbage collection method	Total Average Maximum Minimum
kafka_controller	Offline Partitions	Count	Number of partitions that cannot be written to or read from because there is no active leader	Total Average Maximum Minimum
kafka_controller	Active Controller	Count	Number of active controllers in the cluster	Total Average Maximum Minimum
kafka_controller	Total of Topics	Count	Number of global topics observed in the active controller	Total Average Maximum Minimum
kafka_controller	Leader Elections	Count/Minute	Leader election when a broker fails	Total Average Maximum Minimum
kafka_controller	Unclean Leader Elections	Count/Minute	Elect a broker with OSR (out-of sync replica) as the leader	Total Average Maximum Minimum
kafka_network	Request Queue	Count	Request queue size	Total Average Maximum Minimum
kafka_server	Zookeeper Expired Sessions	Count/Minute	Zookeeper expired session	Total Average Maximum Minimum
kafka_server	Incoming Messages	Count/Second	Message received by the Broker	Total Average Maximum Minimum
kafka_server	Producer requests	Count/Second	Request (Produce) received by the Broker	Total Average Maximum Minimum
kafka_server	Consumer requests	Count/Second	Request (Fetch) received by the Broker	Total Average Maximum Minimum
kafka_server	Failed Fetch Requests	Count/Minute	Number of failed fetches by the Consumer	Total Average Maximum Minimum
kafka_server	Failed Produce Requests	Count/Minute	Number of failed requests by the Producer	Total Average Maximum Minimum
kafka_server	Incoming Byte Rate	Bytes/Seconds	Byte transmission rate received from client (per broker)	Total Average Maximum Minimum
kafka_server	Outgoing Byte Rate	Bytes/Seconds	Byte transmission rate to the client (by broker)	Total Average Maximum Minimum
kafka_server	Rejected Byte Rate	Bytes/Seconds	Bytes rejected per broker because the record batch size exceeds the max.message.bytes setting	Total Average Maximum Minimum
kafka_server	Online Partitions	Count	Total number of partitions	Total Average Maximum Minimum
kafka_server	Leader Partitions	Count	leader partition count	Total Average Maximum Minimum
kafka_server	Under Replicated Partitions	Count	Number of partitions with insufficient replica count	Total Average Maximum Minimum
kafka_server	Under Minisr Partitions	Count	Number of partitions whose in-sync replica count is less than minIsr	Total Average Maximum Minimum
Systemd	Active Systemd Unit	None	Service Status 1 - Service Running 0 - Service Stopped	-

Table. Event Streams basic metrics