The page has been translated by Gen AI.

Overview

Service Overview

ServiceWatch is a service that provides various tools to collect and monitor metrics, logs, and events for resources created in the Samsung Cloud Platform, offering observability across the overall performance and operational status of those resources.

Key Features

It offers the following key features.

Resource Monitoring: Collects and visualizes resource performance metrics (such as CPU Usage). * You can also create a dashboard to visually monitor multiple metrics in one place, providing an at-a-glance overview.
Alert Policy Configuration and Automatic Notifications: You can create alert policies by setting predefined conditions and thresholds. When a threshold is exceeded, you receive notifications, enabling you to quickly check and respond to the status of your resources.
Log Analysis and Storage: Collect logs generated from resources to easily query and search them. * Collected logs are stored and managed in a log group, which provides up to 5 GB of free storage. * You can also configure a log retention policy to specify the retention period. Logs that exceed the retention period do not require separate management.
Cost efficiency: ServiceWatch offers a flexible pay-as-you-go pricing model, allowing you to use it cost-effectively. * It also offers a free tier, allowing you to try it for free and then scale up to a paid plan as needed.

Features

It provides the following features.

Surface Monitoring
- Metrics: ServiceWatch collects and stores metric data provided by services on the Samsung Cloud Platform, and then provides this data to users.
- Dashboard: Visualizes metrics for a single region, providing a unified view of resources. * In ServiceWatch, dashboards are categorized into service dashboards, which are automatically generated for each service, and user dashboards, which are created directly by users.
- Alerts: Provides an alert feature to monitor metric changes based on user-defined thresholds and sends notifications when a threshold is exceeded.
Log Monitoring
- ServiceWatch provides log management features. * Logs collected from the Samsung Cloud Platform service are stored in log groups for management. * You can manage the log retention period by configuring a log retention policy. * You can also query and search log data through the console, and the service provides the ability to save log groups to Object Storage.
ServiceWatch Agent
- You can collect detailed metrics on processes, CPU, memory, disk usage, and network performance from Virtual Servers, GPU Servers, Bare Metal Servers, and more through the ServiceWatch Agent. * You can also collect GPU performance metrics. * You can also collect logs generated from resources through the Agent. * (Scheduled for December 2025)
Event Monitoring
- ServiceWatch creates event rules from system events related to changes in resources created in Samsung Cloud Platform, allowing you to receive notifications when specific conditions are met.

Components

Metric

A metric refers to system performance data. By default, we provide basic monitoring based on free metrics for the resources of services integrated with ServiceWatch. Additionally, services such as Virtual Server can enable detailed monitoring to provide paid metrics.

Metric data is available for up to 15 months (455 days).

For details about metrics, see Metrics.

Log

You can collect, store, and query logs from systems, applications, and services used in Samsung Cloud Platform services, such as Virtual Server and Kubernetes Engine.

For more information about logs, see Logs.

Event

Events indicate changes in the environment within the Samsung Cloud Platform service. The following is an example of an event.

Generates an event when the state of a Virtual Server changes from Stopped to Running.
Creates an event when a new bucket is created in Object Storage.
Generates an event when an IAM user is removed from a user group.

For details about events, see Events.

Dashboard

ServiceWatch provides pre-built service dashboards for each service, and users can also create their own dashboards.

ServiceWatch Agent

ServiceWatch Agent is Virtual Server, GPU Server, and On-Prem. A software component that collects metrics and logs from servers and other sources. This allows you to monitor your infrastructure and applications in greater detail than the default monitoring provided out of the box.

For more information about the ServiceWatch Agent, see Custom Metrics and Logs.

Notice

When collecting custom metrics or logs using the ServiceWatch Agent, contact us separately for the kr-south3 region by referring to Support Center > Inquiries.

Constraints

The limitations of ServiceWatch are as follows.

Category	Description
Metric Query Period	You can set the metric query range to a maximum of 455 days from the query time. This applies to dashboards, widgets, and metric queries.
Number of metrics to query	You can select up to 500 metrics to view them in a graph.
Download indicator image file	You can download image data for up to 100 metrics.
Export Metrics Object Storage	You can export metric data to Object Storage for up to 10 metrics, with a maximum query period of 2 months (63 days).
Number of widgets/metrics per dashboard	Up to 500 widgets per dashboard Up to 500 metrics per widget within a dashboard Up to 2,500 metrics can be added across all widgets in a single dashboard
Number of alert policies	5,000 or fewer per account/region
Alert History	Alert history is available for 30 days.
Number of alert recipients by alert policy	100 or fewer
Number of log groups	10,000 or fewer per account/region
Download Logs	When downloading to Excel, you can download up to 10,000 log events, with a limit of 1 MB per log event. If a log event exceeds 1 MB or the number of log events exceeds 10,000, we recommend using the log group export feature.
Number of log group export jobs	You can run the export job once per account, one at a time.
Log event size	1 MB or less
Number of event rules	300 or fewer per account/region
Event pattern size	2MB or less
Number of notification recipients per event rule	100 or fewer

Table. ServiceWatch limitations

The following is the monthly free provision record for ServiceWatch.

Category	Provided for free
Log	Store up to 5GB per month
Metrics	Default monitoring metrics for each service Detailed monitoring metrics / Custom metrics / Log pattern metrics: 10 per month
Dashboard	Dashboards referencing 50 or fewer metrics: 3 per month Dashboards referencing more than 50 metrics: 1 per month
Alert Policy	10 per month

Table. ServiceWatch free offering scope

Provided Status by Region

ServiceWatch is available in the following environments.

Legion	Availability
Korea West (kr-west1)	Provided
Korea East (kr-east1)	Provided
South Korea 1 (kr-south1)	Provided
South Korea 2 (kr-south2)	Provided
South Korea 3 (kr-south3)	Provided

Table. ServiceWatch availability by region

Prior Service

ServiceWatch has no preceding services.

Release Note

Indicator