1 - Overview

Cloud Monitoring service termination notice

According to Samsung Cloud Platform’s policy, the Cloud Monitoring service is scheduled to be discontinued.
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 in the ServiceWatch Overview. Please refer to it.

※ For some Database and Data Analytics services, refer to the user guide of the respective service for the service watch implementation schedule.

Table. Database, Data Analytics User Guide List

Service Overview

The Cloud Monitoring service collects usage status, change information, and logs of operational infrastructure resources, and generates an event to notify when a configured threshold is exceeded. Through this, users can respond quickly to performance degradation and failures, and can conveniently develop resource capacity expansion plans to configure a stable computing environment.

Provided Features

Cloud Monitoring provides the following features.

  • Stable computing resource management: You can easily view metrics such as CPU usage, disk usage, and memory usage. When an event occurs in the resources being used, an automatic notification is sent to the designated recipients, enabling rapid fault analysis and response, so computing resources can be operated reliably.
  • Convenient Monitoring: Status information about resources can be easily monitored by creating a dashboard. * Provides default and custom dashboards, enabling you to configure various widget types and easily and quickly create dashboards yourself.
  • Event Metric Management: Through the web-based Console, you can easily set event metrics with just a few clicks. The event metric settings for the monitoring target (such as event patterns, trigger conditions, occurrence frequency, performance metrics, operational status, etc.) can be varied to suit the usage environment, and threshold and alarm configurations can be managed conveniently.
  • Resource Log Management: Collects and stores log data of resources, and allows searching the target logs when needed. Additionally, we metricize events for major keywords and automatically notify the designated person when pre‑set conditions are met, providing a more stable usage environment.

Component

Dashboard

In the monitoring dashboard, you can view the operational status and event status of monitored services and resources, as well as the top usage items.

ItemExplanation
RegionResource location
Data reference timeReference time of the data displayed on the dashboard
RefreshRefresh the dashboard based on the current time
Period settingSet data query period and refresh interval
Monitoring statusNumber and status of monitoring targets for each service used in the Account
Event HistoryDisplay events that occurred in the past 7 days as a graph by risk level.
Top 5 usage rates by performanceDisplay the top five monitoring targets with the highest usage for each major performance metric
Event mapDisplay the number of events per service by severity
Event statusDisplay the list of unprocessed events among the occurred events
Table. Cloud Monitoring dashboard configuration

Performance Analysis

In performance analysis, you can identify the main performance metrics of the monitoring target and view the current data and historical records within the period for each metric. Users can view the performance status of the monitoring targets they manage by service or by period, and compare specific performance metrics to analyze the results.

Log Analysis

In log analysis, you collect the logs of the monitoring target, examine their contents, and convert them into metrics—structured data—for monitoring. Each monitoring target provides a default collection log, and users can create custom logs to collect and view additional logs as needed.

Event Management

An event is a configuration that notifies the user when a monitoring target’s performance value meets a specific condition. By configuring events, you can capture essential monitoring information that users need to know without missing it. For example, if you configure events to trigger whenever a performance metric related to overload exceeds a certain threshold, users will receive notifications each time there is a risk of overload during resource operation. Users can proactively respond before problems arise based on this. In event management, you can create such events and configure them to notify designated users whenever a specific value occurs during monitoring.

Preceding Service

Cloud Monitoring has no prerequisite services.

2 - How-to guides

Cloud Monitoring service termination notice

According to Samsung Cloud Platform’s policy, the Cloud Monitoring service is scheduled to be discontinued.
Accordingly, starting after the September 2026 release, monitoring of Samsung Cloud Platform resources through 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.

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

※ For some Database and Data Analytics services, refer to the user guide of the respective service for the service watch implementation schedule.

Table. Database, Data Analytics User Guide List

Samsung Cloud Platform Monitoring is a resource management system that can monitor and analyze the resource operation status within an account operated in the Samsung Cloud Platform Console. Users can efficiently manage resources by using the dashboard page, widgets, and chart features.

reference
  • The user can monitor resources created on an Account with permissions in the Samsung Cloud Platform Console.
  • The user can log in to the Samsung Cloud Platform Console and navigate to Samsung Cloud Platform Monitoring to monitor.

Cloud Monitoring Getting Started

To start Samsung Cloud Platform Monitoring, follow these steps.

  1. All Services > Management > Cloud Monitoring Click the menu. 1. Navigate to the Service Home page of Cloud Monitoring.
  2. On the Service Home page, click the Open Cloud Monitoring button. 2. Go to the Cloud Monitoring Console page.

Explore Cloud Monitoring Console

The top and left menus of the Cloud Monitoring Console are organized as follows.

CategoryDetailed description
Custom Dashboard ManagementCustom Dashboard
  • You can view and manage custom dashboards.
SupportSupport
  • provides links to the user guide and OpenAPI guide
Region ListRegion list
  • Displays the region of the Account currently being monitored.
  • You can select the region provided by the Account
User InformationYou can view user information and log out from Samsung Cloud Platform Monitoring.
Side menuDisplays the main features of Samsung Cloud Platform Monitoring. Clicking each menu takes you to the corresponding page.
  • Monitoring Dashboard: You can view the operational status and event status of monitored services and resources, as well as top usage items. For more details, see 모니터링 대시보드 활용하기.
  • Performance Analysis: You can check the main performance metrics of the monitored target and view the current data and historical data within the period for each metric. For more details, see 성능 분석하기.
  • Log Analysis: You can collect logs of the monitored target, review their contents, and convert them into structured metric data for monitoring. For more details, see 로그 분석하기.
  • Event Management: This is a setting to notify users when performance values meet specific conditions. For more details, see 이벤트 관리하기.
Table. Monitoring page overview

Stop Monitoring

To exit the Cloud Monitoring Console, click the User Info > Logout button at the top right.

reference
The session timeout for the Cloud Monitoring Console is set to 30 minutes.

Using Common Features

This explains the frequently used features when using the Cloud Monitoring Console.

View detailed information of the monitoring target

If you access Cloud Monitoring Console > Performance Analysis or Cloud Monitoring Console > Log Analysis > Log Overview, you can view the list of monitoring targets. At this point, to view detailed information for a monitoring target, click the desired target in the monitoring target list.

Reference
  • Detailed information of the monitoring target varies depending on the service type.
  • If the operating system (OS information) of the monitoring target is RHCOS (Redhat Core OS), detailed information about the monitoring target is not provided.
ItemExplanation
Basic InformationDisplay basic information about the monitoring target
  • Example: Virtual Server - monitoring target, service type, service status, server type, OS information, IP
PerformanceDisplay the primary performance of the monitoring target in a graph
logShow the log collection volume for the monitoring target in a graph.
eventDisplay the list of events that occurred in the monitoring target.
AgentProvides the agent’s install, start, stop, delete, update commands
Set query periodDisplays the reference date/time for data retrieval
  • Refreshes directly to the current time.
  • Turn the automatic refresh feature on or off.
  • You can set the data retrieval period or change the automatic refresh interval. For details, refer to Configure query period
Monitoring status areaDisplays performance, log, and event monitoring status.
Table. Detailed information of the monitoring target
Reference
  • The services that provide agent management commands are Virtual Server, GPU Server, and Bare Metal Server.
  • For detailed information on installing and managing the agent, see Agent Management.

Sorting data

You can organize and view information such as event monitoring, performance, and log analysis results in descending or ascending order. To sort the data, follow these steps.

  1. Display the information to be verified on the page.
  2. Click the Sort button next to the Category name. 2. Each click toggles the sorting order between descending and ascending.

Check real-time data

You can configure the dashboard or detail page data to automatically refresh at a set interval.

Reference
  • In the Cloud Monitoring Console, you can configure whether to enable refresh and set the refresh interval so that the monitoring page refreshes periodically.
  • Click the refresh button to manually refresh based on the current time.

To set the data refresh interval, follow these steps.

  1. Click the Settings button at the top right of the data display area.
  2. After selecting the refresh interval, click the Confirm button.
  3. You can turn the refresh feature on or off.

Configure the query period

By setting the query period, you can limit the query scope to the specified range of performance, logs, and events, making it easy to find only the information you need. To set the query period, follow these steps.

  1. Click the Settings button in the upper right of the data display area.
  2. Select a date range or enter it manually.
Caution
  • If you manually enter the query period, you must set the period to at least 30 minutes.
  • If each widget’s data query range is fixed, the widget’s query range takes precedence.

2.1 - Using Monitoring Dashboards

In the monitoring dashboard, you can view the operational status and event status of monitored services and resources, as well as the top usage items.

Getting Started with Monitoring Dashboard

When you navigate from the Samsung Cloud Platform Console to the Cloud Monitoring Console page, the monitoring dashboard is displayed. If you are on a different page, click Cloud Monitoring Console > Monitoring Dashboard to go to the Monitoring Dashboard page.

The monitoring dashboard is structured as follows.

Itemdescription
Data reference timeDisplay the reference time for the data shown on the dashboard
RefreshRefresh the dashboard based on the current time
Automatic refreshYou can enable or disable the dashboard refresh feature.
Period settingSet the data retrieval period or change the refresh interval
Monitoring StatusDisplay the number of monitoring targets and monitoring status for each service
Event HistoryDisplay the number of events that occurred in the last 7 days as a graph by severity.
Top 5 usage rates by performanceDisplay the usage rates of the five monitoring targets with the highest usage for each major performance metric as a usage graph.
Event mapDisplay the number of events per service by severity
Event statusDisplay the list of unprocessed events among the occurred events.
Table. Monitoring dashboard configuration
reference
  • The monitoring dashboard is automatically created when an Account is created in the Samsung Cloud Platform Console and cannot be deleted arbitrarily.
  • Configuration widgets on the monitoring dashboard cannot be modified arbitrarily.
  • To create a dashboard with a specific widget, use a custom dashboard. For more information about custom dashboards, refer to Using Custom Dashboards.

Explore Common Dashboard Features

This describes the functions available on the dashboard.

Download widget image

Click the download button at the top right of the widget area to download the widget as an image file (*.png).

View graph details

When you place the mouse cursor over the graph, detailed information appears as a popup.

Monitoring Status

Shows the number of monitoring targets and their monitoring status for each service in use.

Itemdescription
Service CategoryDisplay the monitoring target service categories per service and the quantity of monitoring targets included in each service category
  • When you click a service category, the list of services and their quantities included in that category are displayed
Service ListDisplay the list and quantity of services included in the monitoring target service category
  • Click a service’s quantity to go to the Performance Analysis page
Monitoring statusDisplays the number of monitoring targets and their current status
  • Clicking a Down or Unknown item shows the corresponding service name in a popup
Event statusDisplays the number of current events by grade (Fetal, Warning, Inform).
Reference
  • Performance collection in monitoring mode aggregates and displays the number of performance metrics from both Agent and Agentless approaches.

Event History

Displays the number of events that occurred in the last 7 days as a graph by severity.

  • When you place the mouse cursor over the graph, a popup shows the number of occurrences of events corresponding to the selected date’s event risk level, along with active/inactive information.

    • Occurrence: total number of event occurrences
    • Activation: The state where an event that has occurred by meeting the event trigger conditions continues to be maintained.
    • Deactivation: The event that occurred no longer meets the event trigger conditions and has returned to a normal state
  • You can click the risk legend area to hide or unhide the corresponding graph.

Top 5 Usage by Performance

Displays a usage graph for the five monitoring targets with the highest utilization rates across major performance categories.

  • When you place the mouse cursor over the graph, a popup displays the full name of the selected item and its current performance metrics.
  • When you click the graph, a Monitoring Target Details popup window for the corresponding item opens.
    Itemdescription
    CPU Usage/Core [Basic]Percentage of CPU time used, excluding Idle and IOWait states
    Memory Used [Basic]Current memory usage
    Disk Read Bytes [Basic]Disk read byte count
    Disk Write Bytes [Basic]Disk write byte count
Reference
  • The monitoring dashboard only displays the performance of Virtual Server. To show the Top 5 performance of other service types, you need to select and configure them in a custom dashboard.

Event Map

Displays the number of events per service by severity.

  • When you place the mouse cursor over the rectangle, the name of the monitoring target appears as a popup.
  • When you click a service item on the event map, the Monitoring Target Details popup window opens.

The risk level for each item is as follows.

Itemdescription
No RuleThe condition cannot be classified as normal or abnormal. This indicates that the status cannot be assessed due to the absence of a threshold setting.
NORMALIt is in a normal state. This means the threshold did not meet the configured value, so no event was generated.
INFORMThis is the lowest level of risk status, including information at a simple notification level.
WARNINGIt is a moderate risk condition.
FATALThis is the most dangerous stage.

Event Status

Displays a list of events that are in an active state among the generated events.

  • Events are displayed in order of most recent occurrence.

2.2 - Performance Analysis

In performance analysis, you can view the key performance metrics of the monitoring target and check both the current data and historical data within the period for each metric. Users can view the performance status of the monitored targets they manage by service or by period, and compare specific performance metrics to analyze the results.

Getting Started with Performance Analysis

You can start performance analysis by directly selecting the monitoring target or entering search criteria. To search for the monitoring target and analyze performance, follow these steps.

  1. Click Cloud Monitoring Console > Performance Analysis. You will be taken to the Performance Analysis page.
  2. After entering the search criteria for the monitoring target to be analyzed in the search area, click Search.
    Itemdescription
    Search areaThe detailed search filters displayed in the search area vary according to the service type
    • To perform Detailed Search, click the Detailed Search button.
    • Each detailed search filter condition can be selected with one or more items
    Number of monitoring targets displayedDisplay the number of performance items that can be viewed at once in the search results and list
    • The default number of performance items shown in the list is 20 per page.
    • Change the list display count to 10, 20, 30, 40, 50, or 100 per page.
    Search informationDisplay search result values for the search criteria items
    • Monitoring target, service status, event grade
    • Clicking the risk icon displayed for event risk opens a detailed popup of the most recent event corresponding to that risk level.
    Performance metricsInformation Displays key performance indicators according to the service type of the monitoring target
    • The list of key performance indicators per service refers to the service-specific key performance indicators and the collected information by instance type and status of the DB service
    View DetailsView detailed information of the relevant monitoring target
    Performance ComparisonSelect a monitoring target and compare performance
    Table. Performance analysis

Check detailed performance information

To view detailed performance information of the monitoring target, follow these steps.

  1. Click the monitoring target for which you want to view detailed information in the performance analysis list. Monitoring Details popup window opens.
  2. Click the Performance tab.
    • When you place the mouse cursor over the graph, the values of each performance metric appear in a popup window.
    • Click the icon in the upper right corner to set the query period or change the refresh interval.
    • You can click the Details, Summary buttons located at the top left of the performance chart to select the graph display method.
      Itemdescription
      Basic InformationDisplay basic information about the monitoring target
      DetailsPerformance charts of the monitoring target are expanded and displayed
      • View a single chart in detail
      SummaryPerformance charts of the monitoring target are displayed in a grid layout
      • View multiple charts at a glance
      Set query period
      • Date/Time: Displays the reference date and time for data retrieval.
      • Refresh: Manually refresh to the current time.
      • Start/Stop: Turns the automatic refresh feature off or on.
      • Settings: Set the data query period or change the automatic refresh interval
      Performance ComparisonGenerate a chart that compares the performance of monitoring targets, allowing each performance to be compared.
      Performance chartPerformance charts of the monitoring target are displayed as graphs
      • When there is a single graph, the most recent collected value is shown in the upper right corner with its unit.
      • When multiple graphs are present, an ⓘ appears in the upper right corner, and hovering the mouse cursor displays the latest collected value for each graph in a popup.
      • Hovering the mouse cursor over a graph shows the performance metric value at the specified time in a popup.
      Table. Monitoring Target Details
Reference
  • The collection interval of performance metrics may vary depending on the service.
  • The chart displays data at 30 points, and the data collection interval based on the data query range (time) is as follows. (The displayed points may vary due to collection time errors.) 30 minutes: approximately 1‑minute intervals 60 minutes: approximately 2‑minute intervals 3 hours: approximately 6‑minute intervals 6 hours: approximately 12‑minute intervals 12 hours: approximately 24‑minute intervals 24 hours: approximately 48‑minute intervals
    • Day 3: approximately 144-minute interval (2 hours 24 minutes) 7 days: approximately 336-minute interval (5 hours 36 minutes)
    • Day 14: approximately 672‑minute interval (11 hours 12 minutes) Custom: value obtained by dividing the custom range (minutes) by 30
  • The data for each point represents the maximum value within the query range (time), and you can change the statistical type in the detailed chart.

Compare performance

You can view the performance metrics of each monitoring target and select the desired metrics for comparison.

Getting Started with Performance Comparison

Generate a chart that compares the performance of monitoring targets, allowing you to compare each performance.

Reference
  • Only performance metrics of the same service type can be compared.
  • Performance items may be added based on the detailed attributes of the service type.
    • Performance of Windows OS on a VM
    • Search Engine’s Kibana-related performance

To begin the performance comparison, follow these steps.

  1. Click Cloud Monitoring Console > Performance Analysis. You will be taken to the Performance Analysis page.

  2. After entering the search criteria for the monitoring target to be analyzed in the search area, click Search.

  3. After selecting all monitoring targets to compare performance, click Compare Performance. A popup window that allows performance comparison will open.

    Itemdescription
    Monitoring targetDisplay the service type of the monitoring target to compare and click to change the service
    • Changing the service will remove all charts created so far.
    • Click Add to search for monitoring targets of the currently selected service and add
    • The selected monitoring target is displayed on the page, and you can delete the monitoring target by clicking X or Delete All
    Performance itemsDisplay all performance metrics collected from the currently selected service
    • Check the items you want to compare performance for, and those performance items will be included in the chart.
    Chart display methodSelect display method for performance comparison chart
    • Detailed: The performance comparison chart is displayed in detail. (default)
    • Summary: The performance comparison chart is displayed in summary
    Set query period
    • Date/Time: Displays the reference date and time for data retrieval
    • Refresh: Refreshes directly to the current time.
    • Start/Stop: Turns the automatic refresh feature off or on.
    • Settings: Set the data query period or change the automatic refresh interval
    Chart areaCompare the performance of monitoring targets based on the selected performance metric and display it as a chart.

  4. Click Add. A popup window opens where you can add a monitoring target.

  5. After selecting the monitoring target to compare performance, click the Confirm button. If you select Kubernetes Engine, you must also select the sub-type of that service.

  6. Select the performance metrics to compare. The selected metrics will be added to the chart.

Explore the chart

The performance comparison results are displayed as a chart. Users can modify the shape of the generated chart or download it as an image or Excel file.

  1. When you place the mouse cursor over the graph, the performance metric value for the specified time appears as a popup.
  2. Click a target item in the legend area to hide or unhide the corresponding graph.
    Itemdescription
    Statistical methodsSet the statistical method to display in the graph
    • Statistics are displayed in a graph for a period ranging from a minimum of 5 minutes to a maximum of 6 hours.
    • Default, Maximum, Minimum, Average, Total can be selected. Multiple methods can be selected simultaneously, and the selected items are shown in the legend area
    Chart formatSelect the type of graph to display on the chart
    • Line: line graph
    • Stacked Area: area graph
    • Scatter: scatter plot
    Download chartCheck and download the chart’s Raw Data
    • Chart PNG File: Download the chart as an image file (PNG).
    • Chart Excel File: Download the performance item data displayed in the chart as an Excel file. The chart’s displayed data is a dataset automatically collected based on the query range.
    • Raw Excel File: Collect the entire performance item data shown in the chart within the query period and download it as an Excel file.
    Add time series graph widgetAdd the chart to the custom dashboard as a time series graph widget
    • When you click, a popup window for adding a time series graph widget opens.
    DeleteDelete the performance comparison result chart
    Performance Comparison StatusDisplay performance comparison results as a graph
    • When you place the mouse cursor over the graph, the performance comparison status for that time period is shown in a popup window.

2.3 - Log Analysis

In log analysis, you collect the logs of the monitoring target, review their contents, and convert them into structured metrics for monitoring. Each monitoring target provides default collected logs, and users can create custom logs to collect and view additional logs as needed.

Reference
  • To use log analysis, you must first install and operate a log collection agent. For detailed information on installing and operating the log agent, see Managing the Agent.
  • To collect logs from Kubernetes Engine, you must configure log collection in the Samsung Cloud Platform Console.

Getting Started with Log Analysis

You can view the log status list or search for logs to be monitored to check them. To view the log status list, follow these steps.

  1. Cloud Monitoring Console > Log Analysis > Log Overview. Click Log Overview to navigate to the Log Overview page.
  2. After entering the search criteria for the service to be analyzed in the search area, click Log Search.
    • The list of services that match the search criteria and the search information are displayed at the bottom.
    • Click the View Details button for each service to display that service’s detailed log information.
      Itemdescription
      Search areaThe displayed search filters in the search area vary depending on the service type
      • Advanced Search to perform Advanced Search, click the Advanced Search button.
      • You can select one or more condition items for each advanced search filter
      Number of monitoring targets displayedSearch results quantity and the number of items displayed at once in the list
      • The default is 20 items per page.
      • The list display count can be changed to 10, 20, 30, 40, 50, or 100 items per page
      Search informationDisplay the search result values for the search criteria items.
      View DetailsView detailed information of the relevant monitoring target
      Log SearchCombine keywords and queries to search logs and view detailed information
reference
  • If a Virtual Server or Node is connected to the monitoring target, the corresponding status is also displayed in the search information area.
  • The name of the monitoring target can include Korean characters, English letters (both uppercase and lowercase), numbers, and special symbols (-, _, .), and can be up to 100 characters long.
  • When the monitoring target does not have permission, information about the unauthorized target and a permission verification message are displayed in a popup.

Check detailed log information

You can view detailed log entries and log graphs of the monitoring target.

Check log list

You can view detailed log information in the monitoring detail popup window. To view detailed monitoring information for a log, follow these steps.

  1. Cloud Monitoring Console > Log Analysis > Log Overview. Click Log Overview. Log Overview page will open.
  2. Click the log you want to view detailed information for on the Log Status page. The Monitoring Details popup window opens.
  3. Click the Log tab.
    • When you place the mouse cursor over the graph, the values of each log entry appear in a popup window.
    • Click the icon in the upper right corner to set the query period or change the refresh interval.
    • You can select the graph display method by clicking the Details, Summary buttons located at the top left of each log chart.
      ItemExplanation
      Basic InformationDisplay basic information about the monitoring target
      DetailsCharts for each log of the monitoring target are expanded and displayed
      • View a single chart in detail
      SummaryPerformance charts of the monitoring target are displayed in a grid layout
      • View multiple charts at a glance
      Set query period
      • Date/Time: Displays the reference date and time for data retrieval.
      • Refresh: Manually refreshes to the current time.
      • Start/Stop: Turns the automatic refresh feature off or on.
      • Settings: Sets the data query period or changes the automatic refresh interval
      Performance ComparisonCombine keywords and queries to search logs and view detailed information.
      Performance chartCharts for each log of the monitoring target are displayed as graphs
      • When you place the mouse cursor over the graph, the log entry value at the specified time appears in a popup window.

Search logs to verify

You can combine keywords and queries to search logs and view detailed information.

Reference
The presence and frequency of keywords can be converted into metrics, displayed as charts on the dashboard page, or used to set up related events to receive notifications.

To search the logs, follow these steps.

  1. Cloud Monitoring Console > Log Analysis > Log Overview. Click Log Overview. You will be taken to the Log Overview page.

  2. On the Log Overview page, click Log Search. You will be taken to the Log Search page.

    ItemExplanation
    Monitoring targetDisplay the service type of the monitoring target to compare
    • Click the monitoring target list to change the service
    • Changing the service will cause all charts created so far to disappear.
    • Add button to search for and add monitoring targets of the currently selected service
    • The selected monitoring targets are displayed on the page, and you can delete a monitoring target by clicking X or Delete All.
    Search criteriaSet conditions for the logs to be searched
    Set query period
    • Date/Time: Displays the reference date and time for data retrieval.
    • Refresh: Manually refresh to the current time.
    • Start/Stop: Turns the automatic refresh feature off or on.
    • Settings: Set the data query period or change the automatic refresh interval
    Log volume graphWhen you search the logs, the log entries that match the entered criteria are displayed as a chart.
    Generated log messageLog messages from the monitoring target are displayed by time.

  3. Click the Add button. A popup window opens where you can add a monitoring target.

  4. After clicking the monitoring target, select the log file you want to add.

  5. After selecting the log file, click the Confirm button.

  6. After entering the search criteria, click the Search button. The search results will be displayed on the log volume graph and the log messages.

    Itemdescription
    Add indicatorAdd metrics to log search results
    • Use after searching logs
    Execution HistoryCheck the list of search criteria that were recently executed
    • The execution history displays up to the last 20 executed search criteria
    • You can select the desired search history and input it as the current search criteria
    Search fieldSelect the search field
    ConditionSelect search criteria
    • like , !like , = , != , <= , >= , > , < can be selected
    search valueEnter the keyword to search
    Log SearchSelect the operator (AND, OR) for the newly added search condition
    • Displayed only when a new search condition is added
    Add conditionAdd a new search condition

  7. When you search the logs, the log entries that match the entered criteria are displayed as a chart.

    • Log entries are displayed in seconds.
      ItemExplanation
      Log volume graphThe log volume over the selected period is displayed as a graph
      • When you hover the mouse cursor over the graph, the values of each log entry appear in a popup window.
      • Clicking a bar in the graph displays the list of logs for that point in time.
      Set query period
      • Date/Time: Displays the reference date and time for data lookup
      • Refresh: Manually refresh to the current time.
      • Start/Stop: Turn the automatic refresh feature off or on.
      • Settings: Set the data query period or change the automatic refresh interval
      Monitoring targetThe monitoring target list is displayed
      • When you select a monitoring target to view log messages, the log list shows the content
      Log listLog messages generated from the monitoring target are displayed by time
      • Click the button in the log list to view the full message of that log
      • Click Download to save the currently displayed log messages in Excel and TXT file formats

Check log collection status

You can view the main log collection information for the past 7 days as a chart.

  • When you place the mouse cursor over the graph, detailed information appears in a popup window.
  • Only collected logs are aggregated, and logs that have not been collected are not displayed in the status.
Reference
  • When you create an Account, we provide a default virtual capacity of 1 GB to store the collected logs.
  • All logs can be stopped and restarted as needed.

To check the log collection status, click Cloud Monitoring Console > Log Analysis > Log Collection Dashboard.

Itemdescription
Cumulative log volumeDisplay the amount of logs collected from the 1st of each month in GB
  • Show the cumulative usage to date as a percentage of the total allocated virtual capacity.
Log collection volume for the past 7 daysDisplay the amount of logs collected over the past 7 days by service type in a graph
  • The line chart shows the quantity (kb), and the bar chart shows the cumulative usage rate.
  • Click a monitoring target in the legend area to display only that graph
Log occurrence rate by serviceDisplay logs collected over the past 7 days, categorized by service
  • Clicking a bar graph representing each service shows the monitoring target with the highest log collection within that service on the log collection TOP 10 chart.
Log Collection Top 10Display a graph of the top 10 monitoring targets that collected the most logs in the past 7 days within the selected service, based on log occurrence rates by service
  • Click each point on the graph to view detailed information for that log
  • Click a monitoring target in the legend area to display only that graph
  • Click the graph of the target service to navigate to the Log Overview page
reference

To perform monitoring related to logs, you must first install and operate a log collection agent. For detailed information on installing and operating the log agent, refer to 에이전트 관리하기.

  • Accumulated logs are stored up to a maximum of 1 GB. If it exceeds 1 GB, older logs are automatically deleted.

Check indicator configuration status

You can create a metric to display the occurrence count of log patterns as a time series. To view the metric list, click Cloud Monitoring Console > Log Analysis > Metric Configuration Status.

Reference
Metrics converted to time-series data can be set as events or added to a dashboard for real-time monitoring.
Itemdescription
Search areaThe displayed search filters in the search area vary depending on the service type
  • To perform Advanced Search, click the Advanced Search button.
  • You can select one or more condition items for each advanced search filter
Number of monitoring targets displayedDisplay search results
  • The default is 20 per page.
  • Change the list display count to 10, 20, 30, 40, 50, or 100 per page.
Search informationDisplay the search result values for the search criteria items.
AddAdd a new metric
DeleteSelect the metric in the search information and delete it.

Check detailed indicator information

Follow these steps to view detailed information about the indicator.

  1. Cloud Monitoring Console > Log Analysis > Metric Configuration Status. Click Metric Configuration Status. You will be taken to the Metric Configuration Status page.
  2. Indicator Setting Status Click the indicator name to view detailed information on the page. Indicator Details popup window opens.

Add indicator

You can add a new metric to display the desired log data as a time series.

Reference
  • Log metrics can only be set for monitoring targets where the log agent is installed or logs are being collected. For detailed information on installing and operating the log agent, refer to 에이전트 관리하기.

To add a new metric, follow these steps.

  1. Cloud Monitoring Console > Log Analysis > Click Metric Configuration Status. You will be taken to the Metric Configuration Status page.

  2. Indicator Settings Status on the page, click the Add button. Add Indicator popup will open.

  3. Enter indicator name.

    • Metric names can only use English uppercase and lowercase letters, underscores (_), periods (.), and hyphens (-).
    • To distinguish metrics from general performance, the prefix metricfilter. is automatically added and cannot be removed or changed.
      ItemExplanation
      Indicator NameEnter the name of the metric to create
      Monitoring TargetDisplay the service type of the monitoring target to compare
      • Click the monitoring target list to change the service
      • Changing the service will cause all charts created so far to disappear.
      • Click the Add button to search for and add the monitoring target of the currently selected service
      • The selected monitoring target is displayed on the page, and you can delete the monitoring target by clicking X or Delete All
      Search CriteriaSet conditions for the logs to be searched
      Set Query Period
      • Date/Time: Displays the reference date and time for data retrieval
      • Refresh: Refreshes directly to the current time.
      • Start/Stop: Turns the automatic refresh feature off or on.
      • Settings: Allows you to set the data query period or change the automatic refresh interval.
      Log volume graphWhen you search the logs, the log entries that match the entered criteria are displayed as a chart
      Generated log messageLog messages from the monitoring target are displayed by time.
  4. Click the Add button. A popup window opens where you can add a monitoring target.

  5. After clicking the monitoring target, select the log file you want to add.

  6. After selecting the log file, click the Confirm button.

  7. After entering the search criteria, click the Search button. The search results will be displayed on the log volume graph and the generated log messages.

    Itemdescription
    Add indicatorAdd metrics to log search results
    • Use after searching logs
    Execution HistoryCheck the list of search criteria that were recently executed
    • The execution history displays up to the last 20 executed search criteria
    • You can select the desired search history and input it as the current search criteria
    Search fieldSelect the search field
    ConditionSelect search criteria
    • like , !like , = , != , <= , >= , > , < can be selected
    search valueEnter the keyword to search
    operatorSelect the operator (AND, OR) for the newly added search condition
    • Displayed only when a new search condition is added
    Add conditionAdd a new search condition

  8. Click the Confirm button. A new metric will be added with a toast popup message.

Modify indicator search criteria

To modify the indicator’s search criteria, follow these steps.

  1. Cloud Monitoring Console > Log Analysis > Click Metric Configuration Status. Metric Configuration Status page will open.
  2. Indicator Settings Overview On the page, click the indicator name of the metric you want to edit. The Indicator Details popup will open.
  3. In the Indicator Details popup, click the Edit button. The Edit Indicator popup opens.
  4. Metric Update After modifying the search criteria in the popup window, click the Confirm button. The metric will be updated along with a toast popup message.

Delete indicator

To delete the indicator, follow these steps.

reference
  • If there are charts or event policies that use the metric you want to delete, you cannot delete that metric.
  1. Cloud Monitoring Console > Log Analysis > Metric Configuration Status. Click it. You will be taken to the Metric Configuration Status page.
  2. On the Indicator Settings page, select the indicator to delete, then click the Delete button. The indicator will be removed along with a toast popup message.

2.4 - Managing Events

An event is a setting that notifies the user when a performance metric of the monitored target meets a specific condition. By configuring events, users can capture essential monitoring information without missing it. For example, if you set an event to trigger whenever a performance value related to overload exceeds a certain threshold, a notification is sent to the user each time there is a risk of overload during resource operation. Users can respond proactively before problems occur based on this.

In event management, you can create such events and configure them to notify designated users whenever a specific value occurs during monitoring.

Check Event Status

In the Event Status, you can view information about all generated events, related performance metrics, and the history of event notifications delivered to users. To view the Event Status list, follow these steps.

  1. Click Cloud Monitoring Console > Event Management > Event Status. You will be taken to the Event Status page.
  2. On the Event Status page, enter the search criteria for the service whose event status you want to check in the search area, then click the Search button.
    Itemdescription
    Search areaThe search filters displayed in the search area differ according to the service type
    • To perform an Advanced Search, click the Advanced Search button.
    • You can select one or more condition items for each advanced search filter
    Number of monitoring targets displayedDisplay the quantity of search results and the number of items that can be viewed at once in the list
    • The default number of items shown in the list is 20 per page.
    • The list display count can be changed to 10, 20, 30, 40, 50, or 100 items per page
    Search informationDisplay search result values for the search criteria items
    • Clicking the message content of each service allows you to view detailed event information
    View DetailsView detailed information of the relevant monitoring target
    Table. Event List
Reference
  • If a Virtual Server or Node is connected to the monitoring target, the corresponding status is also displayed in the search information area.
  • The name of the monitoring target can include Korean characters, English letters (both uppercase and lowercase), numbers, and special symbols (-, _, .), and can be up to 100 characters long.

View event status list

In the monitoring detail popup, you can view the event information, occurrence time, and duration in the event list. To check the event occurrence status, follow these steps.

  1. Click Cloud Monitoring Console > Event Management > Event Status. You will be taken to the Event Status page.
  2. On the Event Status page, click the Event tab.
    Itemdescription
    Event statusCheck event message and occurrence time
    activeShow only events that are currently active
    AllShow all events
    Event DetailsCheck the detailed information of the selected message in the event status
    Table. Event tab

Check event details

To view the event details, follow these steps.

  1. Click Cloud Monitoring Console > Event Management > Event Status. You will be taken to the Event Status page.
  2. On the Event Status page, click the Event tab.
  3. On the Event Status page, after selecting the event for which you want to view detailed information, click Event Details to view the event publishing conditions, performance items, and notification history.
    ItemExplanation
    Monitoring targetDisplay the name of the monitoring target
    Occurrence conditionDisplay the condition under which the event occurs
    Performance itemsDisplays a chart for performance items.
    • Hovering the mouse cursor over the graph shows detailed performance values for each time period.
    Notification HistoryDisplay the full alarm occurrence history
    Event Settings DetailsView the configuration information of the event
    Table. Event Details

Manage Event Settings

You can configure event details such as the monitoring target, performance metrics that define the event trigger, event severity level, and event notification recipients. When data collected from the monitoring target meets the conditions set in the event policy, notifications are delivered to the user via email, SMS, or messages.

Reference
  • Event policies can be set only when a monitoring target is specified, and policies for each Auto-Scaling Group can be configured at the group level.

Check Event Settings

To verify the event settings, follow these steps.

  1. Click Cloud Monitoring Console > Event Management > Event Settings. Navigate to the Event Settings page.
  2. On the Event Settings page, enter the search criteria for the service whose event policy you want to check in the search area, then click the Search button.
    ItemExplanation
    Search areaThe search filters displayed in the search area vary depending on the service type
    • To perform Advanced Search, click the Advanced Search button.
    • You can select one or more condition items for each advanced search filter
    Number of monitoring targets displayedDisplay search results
    • The default is to show 20 items per page.
    • Change the number of items displayed in the list to 10, 20, 30, 40, 50, or 100 per page.
    Monitoring targetDisplay the name of the monitoring target
    • When the checkbox is selected, the Delete, Activate, and Notification Recipient buttons become enabled.
    Performance itemsDisplay performance items for the event configuration target
    Individual itemDisplay individual performance items under the performance category
    • If there are no individual items, nothing is displayed.
    Type/UnitDisplay the value type and unit of the performance item
    Event ratingDisplay the risk level of the event
    • The risk level is set manually by the user when adding an event
    • Fatal: The highest risk level.
    • Warning: A medium-level risk.
    • Information: The lowest risk level, for reference.
    thresholdDisplay the reference value for comparing performance values.
    Notification recipientsDisplay the recipients of the event notification
    • When the mouse cursor is placed over the name, the full list is displayed on the page
    Policy statusIndicates whether the event is active
    View DetailsCheck and edit event details
    • Click ‘View Details’ to open the detailed information popup for the event.
    AddAdd event
    DeleteDelete event
    EnableEnable or disable the event
    Notification recipientsCheck and manage event notification recipients
    Table. Event Settings
Reference
  • The name of the monitoring target can include Korean characters, English letters (both uppercase and lowercase), numbers, and special symbols (-, _, .), and can be up to 100 characters long.
  • When the monitoring target does not have permission, information about the unauthorized target and a permission verification message are displayed in a popup.

Check detailed event settings

You can view detailed information about the monitoring target and event conditions, and modify the event conditions and notification information.

Add Event Settings

To add an event setting, follow these steps.

Reference
  • Event policies can only be set when a monitoring target is specified.
  • Auto-Scaling Group policies can be applied on a per-group basis.
  1. Click Cloud Monitoring Console > Event Management > Event Settings. Navigate to the Event Settings page.

  2. Click the Add button on the Event Settings page. The Add Event Settings popup opens.

    Itemdescription
    Target NameSelect the monitoring target to add an event setting
    • Click the monitoring target list to change the service
    • Changing the service will delete all event conditions created so far.
    • Click the Add button to search for and add the monitoring target of the currently selected service
    • The selected monitoring target is displayed on the page, and you can delete the monitoring target by clicking X or Delete All
    Event Settings AreaSet the performance and occurrence conditions for the event
    Notification information areaSet the notification recipients and method when an event occurs.
    Table: Description of the Add Event Settings Popup

  3. After selecting the service type in the monitoring target area, click the Add button. The add monitoring target popup window will open.

  4. After selecting the monitoring target, click the Confirm button.

    • You can select multiple monitoring targets simultaneously.
    • If there are multiple monitoring targets, the configured event is added identically to each monitoring target.
    • If you select Kubernetes, you must also select the sub-type of that service.
  5. In the performance items, click the item where you want to add an event, then enter the event trigger condition.

    • The added performance items display the count of additions next to the performance name.
    • If you select multiple performance items, you must enter the event occurrence condition for each performance item.
      Itemexplanation
      Load Event Policy TemplateSelect and apply an existing event policy template.
      Performance itemsClick the performance metric for which you want to set the event trigger condition and add it to the event condition configuration area.
      Event ratingSet the event severity level
      • Fatal: the most dangerous level.
      • Warning: a medium-level risk.
      • Information: the lowest level of risk and for reference.
      Performance typeSelect the reference value for determining whether an event occurs
      • Collected value: Use the current value.
      • Delta value: Use the difference between the previous and current values.
      thresholdSet the reference value to compare with the collected performance values
      • It serves as the criterion for determining whether an event occurs.
      • Only numbers and decimal points can be entered
      Comparison methodSelect a method that compares the monitoring value of the performance item with the threshold to determine whether an event occurs
      • Range: Check whether the performance value is within the range specified by the threshold
      • Match: Check whether the performance value equals the threshold
      • Different: Check whether the performance value differs from the threshold
      • At least: Check whether the performance value is greater than or equal to the threshold
      • Greater than: Check whether the performance value exceeds the threshold
      • At most: Check whether the performance value is less than or equal to the threshold
      • Less than: Check whether the performance value is less than the threshold
      Individual itemSpecify an individual performance item under the performance items as an event condition
      • It is enabled only when the performance item can collect the individual item.
      PrefixYou can add a prefix to the event message.
      • Event Status page uses this as a keyword to search for the event.
      StatisticsSet the statistical method to apply to the collected performance values
      • When statistics are configured, the performance value calculated using the selected statistical method is compared against the threshold when evaluating event trigger conditions. If not selected, the most recent performance value is compared to the threshold.
      • Statistical Method: Choose one of maximum, minimum, average, or sum to compute the collected performance values.
      • Statistical Period: Set the time span over which the statistical method is applied. It is the period measured from the most recently collected performance value.
      Sustained occurrence countSet the number of consecutive monitoring values that satisfy the event occurrence condition
      • This value is used as a sensitivity to determine whether the event is a momentary outlier or a real event.
      Event occurrence notification time zoneTimezone setting feature when configuring event policies
      Table. Add Event Settings - Event Settings Area
  6. Notification area allows you to set notifications.

    Itemdescription
    Notification recipient selection areaSelect notification recipients
    • After selecting the notification recipients, click the Delete button to remove the selected recipient.
    Notification recipients / groupThe list of recipients to receive the notification when an event occurs is displayed.
    Event risk levelThe risk level of the configured event is displayed.
    Notification methodThe method of delivering notifications to the recipient is displayed.
    AddSelect and add a new notification recipient from the address book.
    DeleteDelete a notification recipient from the notification recipient/group
    Table. Add event settings - notification info area

  7. Check the notification recipients, select them, and click the Confirm button.

reference
  • Only the account’s root user or an IAM user can be added as a notification recipient.
  • You can select multiple items simultaneously.
  1. Set the notification method for each recipient according to the event risk level.
    • The notification method can be selected from email, SMS, and messenger, and multiple methods can be selected simultaneously.
  2. When the notification method setup is complete, click the Confirm button.

Modify Event Settings

To edit the event’s conditions and notification recipient information, follow the steps below.

  1. Click Cloud Monitoring Console > Event Management > Event Settings. You will be taken to the Event Settings page.
  2. Event Settings page, enter the search criteria for the service whose event settings you want to modify in the search area, then click the Search button.
  3. From the event policy list, click the View Details button of the event policy you want to edit. You will be taken to the Event Settings Details page.
  4. On the Event Settings Detail page, click the Edit button. You will be taken to the Event Settings Edit page.
  5. On the Edit Event Settings page, enter the information to be modified, then click the Confirm button.
    • You can edit the event conditions and notification information.

Delete Event Settings

To delete the event configuration, follow these steps.

  1. Click Cloud Monitoring Console > Event Management > Event Settings. You will be taken to the Event Settings page.
  2. On the Event Settings page, enter the search criteria for the service whose event policy you want to delete in the search area, then click the Search button.
  3. After checking the event policy to delete in the event policy list, click the Delete button.
  4. In the confirmation popup, click the Confirm button.

Change Event Settings Activation

You can easily change whether the event policy is enabled.

  1. Click Cloud Monitoring Console > Event Management > Event Settings. Go to the Event Settings page.
  2. On the Event Settings page, enter the search criteria for the service whose event policy you want to delete in the search area, then click the Search button.
  3. In the event policy list, check the event policy whose activation you want to change, then click the Activate button. The Policy Activation popup window will open.
  4. After selecting the activation status, click the Confirm button.
    • Enable All, Disable All buttons can be clicked to change them in bulk.
Reference
If you disable the event policy, all active events generated by the selected event policy will be disabled.

Change Event Notification Recipients

You can verify the recipients of notifications when an event occurs and change them in bulk.

Reference
  • The event notification recipient change feature is intended to modify event notification recipients in bulk. Consequently, the existing recipients are removed and replaced with the new recipient settings.
  • To view and modify the notification recipients for each policy, click the Edit button on the policy’s detail page, then make the changes.
  1. Click Cloud Monitoring Console > Event Management > Event Settings. Go to the Event Settings page.
  2. On the Event Settings page, enter the search criteria for the service whose event policy you want to delete in the search area, then click the Search button.
  3. After checking the event policy to edit in the event policy list, click the Notification Recipients button. You will be taken to the Notification Recipients page.
  4. On the Notification Recipients page, after selecting the user to add as a notification recipient, click the Confirm button.
    Itemdescription
    Event policy listThe list of event policies for changing the notification recipients is displayed
    • Click Add to add the policy to be changed
    • Click the Delete button in the policy list to remove that policy.
    User search areaEnter name, email, mobile phone, and company name to search
    Notification address bookUse the notification address book to verify and add users.
    Search User ListThe list of users included in the notification address book or search results is displayed
    • If you check the users to add as notification recipients, they will be added to the notification recipient list.
    Notification recipient listThe list of users to be added as notification recipients for the event displayed in the event policy list is shown
    • After checking a user, click the Delete button to remove that user from the list.
    Table. Change Event Notification Recipients

Managing Event Templates

You can set the monitoring target, performance values that define event occurrence criteria, and the event risk level, then create and use a template. When adding or modifying an event, you can import an event policy template to easily enter the event conditions.

Check the list of event policy templates

To view the list of event policy templates, follow these steps.

  1. Click Cloud Monitoring Console > Event Management > Event Settings. Navigate to the Event Settings page.
  2. On the Event Settings page, click Event Policy Template. You will be taken to the Event Policy Template page.
  3. On the Event Policy Template page, enter the search criteria for the service whose template you want to check in the search area, then click Search.
    Itemdescription
    Search areaEnter the conditions of the event policy template to search.
    Add event policy templateAdd event policy template
    Template ListEvent policy templates that match the search criteria are displayed.
    Table. Event Policy Template List

Add event policy template

To add an event policy template, follow these steps.

  1. Click Cloud Monitoring Console > Event Management > Event Settings. Navigate to the Event Settings page.

  2. On the Event Settings page, click the Event Policy Template button. You will be taken to the Event Policy Template page.

  3. On the Event Policy Template page, click the Add Event Policy Template button. The Add Event Policy Template popup opens.

  4. Add Event Policy Template In the popup window, set the service type and template information for adding the event policy template. * Items marked with * are required fields and must be entered.

    Itemdescription
    Service TypeSelect the service type to set the event policy
    • Click the service type list to change the service
    • If you change the service, all event conditions created so far will be lost.
    Template nameEnter the name of the template to create
    Template descriptionEnter a description for the template to be created
    Table. Add event policy template – set service type and template name

  5. In the performance items, click the item where you want to add an event, then enter the event trigger condition.

    • The added performance items display the count of additions next to the performance name.
    • If you select multiple performance items, you must enter the event trigger condition for each item. * Items marked with * are required fields and must be entered.
      Itemdescription
      Load Event Policy TemplateSelect and apply an existing event policy template
      • When you load the template, the event conditions and notification recipients are replaced with the information set in the template.
      Performance itemsClick the performance metric to set the event trigger condition and add it to the event condition configuration area.
      Event ratingSet the event severity
      • Fatal: The most dangerous level.
      • Warning: A medium-level risk.
      • Information: The lowest risk level, for reference only.
      Performance typeSelect the reference value to determine whether an event occurs
      • Collected value: Use the current value.
      • Delta value: Use the difference between the previous and current values.
      thresholdSet the reference value to compare with the collected performance values
      • It serves as the criterion for determining whether an event occurs.
      • Only numbers and decimal points can be entered
      Comparison methodTo determine whether an event occurs, select the method that compares the monitoring value of the performance item with the threshold.
      • Range: Check if the performance value is within the range specified by the threshold.
      • Match: Check if the performance value matches the threshold.
      • Different: Check if the performance value differs from the threshold.
      • AtLeast: Check if the performance value is greater than or equal to the threshold.
      • Exceeds Check if the performance value exceeds the threshold.
      • AtMost: Check if the performance value is less than or equal to the threshold.
      • LessThan: Check if the performance value is less than the threshold.
      Individual itemSpecify an individual performance item under the performance items as an event condition
      • It is enabled only when the performance item can collect the individual item.
      PrefixAdd an event message prefix
      • It is used as a keyword to search for this event on the event status page.
      StatisticsSet the statistical method to apply to the collected performance values
      • When statistics are set, the performance value calculated using the configured statistical method is compared to the threshold when determining event trigger conditions. If not selected, the most recent performance value is compared to the threshold.
      • Statistical Method: Choose one among maximum, minimum, average, sum to calculate the collected performance values.
      • Statistical Period: Set the period over which the statistical method calculation is applied. It is the period from the most recently collected performance value.
      Number of occurrencesSet the number of consecutive monitoring values that satisfy the event trigger condition
      • This value is used as a sensitivity to determine whether the event is a transient anomaly or a genuine event.
      Event occurrence notification time windowTimezone setting feature when configuring event policies
      Table. Add event policy template – performance item
  6. Set the recipients and delivery method for the information when a notification occurs.

    Itemdescription
    AddSelect and add a new notification recipient from the address book.
    DeleteDelete the selected notification recipient(s) from the notification recipients/group
    Notification recipients / groupsThe list of recipients to receive the notification content is displayed when an event occurs
    • After selecting a notification recipient, clicking the Delete button removes that recipient.
    Event risk levelThe risk level of the event to be delivered is displayed
    Notification methodThe method of delivering notifications to the recipient is displayed
    • you can choose among email, SMS, and messenger, and you can select multiple methods simultaneously
    Table. Add event policy template – set notification recipients

Reference
  • Only Account members and the notification address book registered in the Account can be added as recipients.
  • You can select multiple items simultaneously.
  1. Click the Confirm button. The event policy template will be added along with a toast popup message.

Edit and delete event policy templates

To modify or delete an event policy template, follow the steps below.

  1. Click Cloud Monitoring Console > Event Management > Event Settings. You will be taken to the Event Settings page.
  2. On the Event Settings page, click the Event Policy Template button. You will be taken to the Event Policy Template page.
  3. On the Event Policy Template page, enter the search criteria for the service whose template you want to view in the search area, then click the Search button.
  4. Click the More button at the top right of the template you want to edit or delete, and then click Edit or Delete.
    • Edit: The template edit popup window opens. After editing the template, click the Confirm button.
    • Delete: The template will be deleted along with a toast popup message.
  5. Click the Confirm button. The template will be deleted along with a toast popup message.

Share event policy template

To share the event policy template, follow these steps.

  1. Click Cloud Monitoring Console > Event Management > Event Settings. Navigate to the Event Settings page.
  2. On the Event Settings page, click the Event Policy Template button. You will be taken to the Event Policy Template page.
  3. On the Event Policy Template page, enter the search criteria for the service whose template you want to view in the search area, then click the Search button.
  4. Click the More > Share button located at the top right of the template you want to share.
  5. After selecting the user to share with, click the > button. The selected user will be added to the sharing target.
  6. Click the Confirm button. The template will be shared with a toast popup message.

Filtering events

You can filter notifications for events that occur during a specific period. While event filtering is applied, notifications will not be delivered even if events occur.

To view the event filtering list, follow these steps.

  1. Cloud Monitoring Console > Event Management > Event Filtering click. You will be taken to the Event Filtering page.
    Itemdescription
    Filtering TimelineDisplay the timeline of registered filters by date
    • Registered filters are displayed on the timeline as bars. Clicking a bar shows the filter’s detailed information.
    • The numbers from 00 on the left to 23 on the right represent the hour of the day.
    • The blue vertical line below the time indicates the current time.
    • < , > Click to change the displayed date
    Filtering listDisplays a list of information and operational status of registered filters
    • Running: The filter is registered and currently operating
    • Ended: The filter’s operation has ended after the set period has passed.
    • Scheduled: The filter registration is complete and is pending. The filter will operate when the set period arrives.
    • Disabled: The filter is in a stopped state. It is displayed when ‘Use’ is not selected in the detailed settings
    AddAdd new event filtering
    DeleteDelete the selected event filter from the filter list
    Search areaSearch for event filtering or monitoring targets
    Table. Event Filtering List
Reference
The filtering timeline chart is displayed based on the time zone set in the logged-in user’s account.

Add event filtering

To add event filtering, follow these steps.

  1. Click Cloud Monitoring Console > Event Management > Event Filtering. Navigate to the Event Filtering page.
  2. Click the Add button on the Event Filtering page. The Add Event Filtering popup opens.
  3. Add Event Filtering Enter the filtering information in the popup window.
    ItemExplanation
    Event filteringEnter the name of the event filter
    Usage statusSet whether event filtering is used
    • If set to Not Used, it will be displayed as Disabled until changed to Enabled, and filtering will not operate.
    Time zoneSet the reference time zone for applying event filtering
    Iteration typeSet the repeat application of event filtering
    • No repeat: Enter the start and end year, month, day, hour, minute. Filtering occurs only once without repetition.
    • Daily, weekdays: Enter only the start time and end time. Filtering repeats daily at the specified times.
    PeriodSet the period during which event filtering is applied
    • Applied Time: For recurring tasks, it is active and displays the elapsed time from the start time to the end time
    • Conversion Period: The event filtering period is converted and displayed based on the time zone set by the user
    Event filtering targetSelect the service type and monitoring target to apply event filtering, then add.
    Table. Add event filtering
  4. Click the Confirm button. Event filtering will be added with a toast popup message.
Reference
You can change whether event filtering is enabled by editing the filter.

Modify Event Filtering

To modify event filtering, follow these steps.

  1. Click Cloud Monitoring Console > Event Management > Event Filtering. Proceed to the Event Filtering page.
  2. Event Filtering page, click the name of the filter you want to edit. Event Filtering Details popup window opens.
  3. Event Filtering Details in the popup window, click the Edit button. The Edit Event Filtering popup window opens.
  4. Edit Event Filtering After entering the changes in the popup window, click the Confirm button. The event filtering will be updated with a toast popup message.

Delete event filtering

To delete event filtering, follow these steps.

  1. Click Cloud Monitoring Console > Event Management > Event Filtering. Navigate to the Event Filtering page.
  2. On the Event Filtering page, select the event filter you want to delete, then click the Delete button. The event filter will be deleted along with a toast popup message.
    • You can select multiple event filters simultaneously.

Managing Notification Groups

You can group the recipients who receive notifications when an event occurs into a single group for management. Notification Group allows you to efficiently manage notification recipients and configure notification settings easily and quickly.

To check the notification group, follow the steps below.

  1. Click Cloud Monitoring Console > Event Management > Alert Group. Go to the Alert Group page.
  2. Notification Group page allows you to view and manage notification groups.
    Itemdescription
    Add notification groupAdd a new notification group.
    Notification GroupDisplays a list of all notification group created by the user.
    • When a notification group is clicked, the notification group details popup opens.
    • Click the Edit button to modify the notification group
    Advanced SearchYou can search the address book by entering the notification group name.
    Keyword searchYou can search by selecting the notification group, user name, creation timestamp, and last modified timestamp.
Reference
Notification Group is only valid within an Account, so it can be composed only of users who have access rights to that Account. Users whose access rights have been removed are automatically excluded from the address book.

Add Notification Group

To add a notification group, follow these steps.

  1. Cloud Monitoring Console > Event Management > Add Alert Group Click.
  2. Add Notification Group page allows you to enter the notification group name and description, then add users.
  3. Click the Save button to add the notification group.

Edit Notification Group

You can add a user to a notification group or delete a user registered in the notification group.

Add User

To add a user to the notification group, follow these steps.

  1. Click Cloud Monitoring Console > Event Management > Alert Group.
  2. In the All Notification Groups, click the notification group to which you want to add a user, then click Edit.
  3. Please select the user to add.
    • Only users registered in the Account can be added to the address book.
    • You can quickly find the desired members using the real-time search GUI.
  4. Click the Save button. The user address will be added with a toast popup message.

Delete Notification Group

To delete a notification group, follow these steps.

  1. Click Cloud Monitoring Console > Event Management > Alert Group.
  2. Click the notification group you want to delete from the overall notification groups.
  3. After selecting the notification group to delete, click Delete.
    • You can select multiple addresses simultaneously.
  4. Click the Confirm button. The address will be deleted along with the toast popup message.

2.5 - Using Custom Dashboards

Custom dashboards are personalized dashboards that users configure by selecting the widgets they want. Users can use custom dashboards to arrange monitoring information as they wish, and they can share the created custom dashboards with other users.
The content covered in Using Custom Dashboards is as follows.

reference
Custom dashboards are created separately from the Account dashboard and can display monitoring information from multiple Accounts at once.

Getting Started with Custom Dashboard

After creating a custom dashboard, the user can add desired widgets to view monitoring information.

Create a custom dashboard

To create a custom dashboard, follow these steps.

  1. From the top‑right menu, click Custom Dashboard Management. You will be taken to the Custom Dashboard Management page.
  2. Click Add Dashboard. The Add Dashboard popup window opens.
  3. Enter the dashboard name to create and click the Save button.
  4. The custom dashboard you created appears in the My Dashboard list.

Add widget

Custom dashboards provide widgets in various formats such as performance statistics, comparison charts, and event lists. Users can add the information they want to monitor as widgets and freely configure the custom dashboard.

Reference
  • You can change the position and size of a created widget, or edit, copy, and delete its content. For more details, see Custom Widget Management.

To add a widget, follow these steps.

  1. From the top‑right menu, click Custom Dashboard Management. You will be taken to the Custom Dashboard Management page.
  2. From the My Dashboard list, select the custom dashboard to add a widget.
  3. Click the + button or the Add Widget button at the top right of the dashboard. The Add Widget popup window opens.
  4. Add Widget In the popup window, select the widget to use on the dashboard and add it.
    • When you select a widget, detailed settings and a preview are displayed.
    • For explanations and configuration methods for each chart, refer to Custom Widget.
  5. Click the Confirm button.
Reference
A widget is added to the dashboard at its default size.

Custom widget

The types of widgets that can be added to a custom dashboard are as follows.

Widget NameExplanation
Title BoxDisplay the title box on the custom dashboard.
Event statusDisplays the event that occurred.
Monitoring StatusDisplays the number of monitoring targets and the monitoring status.
Top 5 Key PerformanceDisplays the top 5 monitoring targets with the highest utilization for a specific performance metric.
Event mapDisplays the number of events per service by severity level.
Event HistoryDisplays the count of events per date by severity.
time series graphDisplays the performance metrics of the selected monitoring target as a time-series graph.
Current status indicatorDisplays the performance value statistics and risk levels of the selected monitoring targets.
Instance mapDisplays the performance values of the selected monitoring targets using colors of varying intensity.
Table. Types of widgets that can be added to a custom dashboard

Title Box

Displays a title box on the custom dashboard.

  • You can create up to 10 title boxes.
  • You can add multiple title boxes at the same time.
Itemdescription
TitleEnter the text to display in the title box.
AddAdd a new text box.
DeleteDelete the corresponding text box.
Table. Custom Dashboard Title Box

Event status

Displays the occurred event.

  • You can configure it to display all events that have occurred, or only the active events.
Itemdescription
Widget nameEnter the name of the widget.
Query rangeSelect the range of events to display in the widget
  • All events: Displays all events that have occurred. Completed events are marked as Completed
  • Unaddressed events: Displays events that have not been addressed so far.
Table. Event Status

Monitoring Status

Displays the number of monitoring targets and the monitoring status.

Itemexplanation
Widget nameEnter the name of the widget.
Table. Monitoring Status

Key Performance Top 5

Shows the top five monitoring targets with the highest usage rate for a specific performance metric within the Account.

Itemdescription
Widget nameEnter the name of the widget.
ServiceSelect the service to check performance.
Performance metricsSelect the performance metric that serves as the basis for displaying the monitoring target
  • CPU Usage/Core [Basic]: Displayed based on CPU usage.
  • Memory Used [Basic]: Displayed based on memory usage.
  • Disk Read Bytes [Basic]: Displayed based on disk read usage.
  • Disk Write Bytes [Basic]: Displayed based on disk write usage.
Table. Top 5 Key Performance

Event map

Displays the number of events per service by severity level.

Itemdescription
Widget nameEnter the name of the widget.
Table. Event Map

Event History

Displays the number of events per date, grouped by severity.

Itemdescription
Widget nameEnter the name of the widget.
Table. Event History

Time Series Graph

Displays the performance metrics of the selected monitoring target as a time-series graph.

  • You can change the period displayed by the time series graph using the dashboard’s date range setting feature.
  • When you place the mouse cursor over the graph, you can view the time and performance values for each target at that point.
Itemdescription
Widget nameEnter the name of the widget.
ServiceSelect the service to check performance.
Monitoring targetSelect the monitoring target to display as a graph.
Performance itemsSelect the performance metric to display as a graph.
Add optionYou can display a danger zone.
  • After selecting the danger zone display checkbox and entering the interval values, the corresponding zone is shown as a red area on the graph.
Table. Time series graph
Reference

You can click the icon at the top right of the preview to change the graph type.

  • Linear graph
  • area chart
  • stacked bar chart
  • scatter plot

Current Indicator

Displays statistical figures and risk levels for the performance values of monitored entities.

In the monitoring dashboard, if you place the mouse cursor over a status indicator value, you can view detailed information for that item.

Itemdescription
Widget nameEnter the name of the widget.
ServiceSelect the service to check performance.
Monitoring targetSelect the monitoring target to display as a graph.
Performance metricsSelect the performance items to display as a graph.
StatisticsSelect the statistical method to display the performance values of the monitoring target
  • avg: Displays the average of all collected performance values.
  • min: Displays the smallest value among all collected performance values.
  • max: Displays the largest value among all collected performance values.
  • raw: Displays the most recent performance value. Use only when a single monitoring target is selected.
Add optionYou can display a danger zone.
  • After selecting the danger zone display checkbox and entering the range values, the specified range will be shown as a red area on the graph.
Table. Status Indicator

Instance Map

Display the performance values of monitoring targets using colors of varying intensity.

  • When you position the mouse cursor over each heatmap, you can view detailed information about the item.
Itemdescription
Widget nameEnter the name of the widget.
ServiceSelect the service to check performance.
Monitoring targetSelect the monitoring target to display as a graph.
Performance metricsSelect the performance metric to display as a graph.
Table. Instance Map

Check Custom Dashboard

To view the custom dashboard, follow these steps.

  1. From the top‑right menu, click Custom Dashboard Management. You will be taken to the Custom Dashboard Management page.
  2. From the My Dashboard list, select the Custom Dashboard.
    Itemdescription
    Dashboard ListDisplays the list of custom dashboards. Click a list item to change the dashboard to view.
    • My Dashboards: Displays the list of dashboards you created yourself.
    • Shared Dashboards: Displays the list of dashboards shared with you.
    Dashboard nameThe name of the user dashboard is displayed.
    Dashboard Settings
    • Date/Time: Displays the reference date and time for the analysis information.
    • Refresh: Refreshes to the current time.
    • Stop/Start: Turns the automatic refresh feature off or on.
    • Settings: You can set the data query period or change the automatic refresh interval. (See “Setting the query period” for reference)
    Add widgetAdd a new widget to the dashboard.
    Dashboard editingYou can edit the currently configured custom dashboard.
    • Dashboard Edit: Modify the name of the currently selected dashboard.
    • Dashboard Copy: Copy the currently selected dashboard to create a custom dashboard with the same widgets.
    • Dashboard Delete: Delete the currently selected dashboard.
    • Dashboard Share: Share the dashboard so that specific users can view it. For more information, see Sharing Custom Dashboards.
    Custom widgetDisplays the widgets that make up the dashboard.
    • You can change the widget’s position and size, or edit and delete it. For more information, see Managing Custom Widgets
    • You can download graphic widgets as image files.
    Table. Custom dashboard information
Reference
You can click the star icon next to the dashboard name to add it to favorites. Dashboards added to favorites are displayed at the top of the dashboard list.

Download widget

Graphic widgets can be downloaded as image files (*.png).
When you hover the mouse over a graph widget, a download button appears in the upper right corner. Clicking the download button downloads the widget as an image file.

Share Custom Dashboard

You can share a custom dashboard and configure it so that other users can view it.

Reference
Dashboard recipients remain as shared recipients even if they are later removed from the current account.

To share a custom dashboard, follow these steps.

  1. From the top‑right menu, click Custom Dashboard Management. You will be taken to the Custom Dashboard Management page.
  2. From the My Dashboard list, select the Custom Dashboard.
  3. Click Dashboard More at the top right, then click Dashboard Share. The Dashboard Share popup opens.
  4. After selecting the user to share the dashboard with, click the > button and verify that the selected user moves to the shared target.
  5. Click the Confirm button.

Managing Custom Dashboards

You can edit, copy, or delete a custom dashboard.

  1. From the top-right menu, click Custom Dashboard Management. You will be taken to the Custom Dashboard Management page.
  2. From the My Dashboard list, select the Custom Dashboard.
  3. Click the Dashboard top-right more button, then select the desired command.
    • Edit Dashboard: Edit the dashboard name.
    • Dashboard Copy: Copy the dashboard to create a new dashboard.
    • Dashboard Sharing: Share the dashboard with other users.
    • Delete Dashboard: Deletes the dashboard.

Managing custom widgets

You can change the widget’s position and size, or edit and duplicate the widget.

Change Widget Position

Click the widget’s name, then drag to change its position.

Changing Widget Size

To change the size of the widget, follow these steps.

  1. Place the mouse cursor over the widget. The size adjustment button appears in the lower right corner of the widget.
  2. Size Adjustment button, click and hold while dragging to adjust to the desired size.

Edit, copy, delete widget

To modify, copy, or delete a widget, follow these steps.

  1. Place the mouse cursor over the widget. The More button appears in the top right corner of the widget.
  2. After clicking the More button, click the desired command.
    • Widget Edit: Modify the widget’s chart settings.
    • Widget Copy: Copies the widget to create a widget with identical content.
    • Delete Widget: Deletes the widget.

2.6 - Managing Agents

The agent is a module that collects performance metrics, logs, and Windows events from the monitoring target. Users must verify the agent’s installation status and operate and manage it in order to use the monitoring functionality.

Caution
  • If IP access control is configured for the monitoring target, you cannot use agent management. If agent management cannot be used, check the IP access control configuration status of the selected monitoring target.
  • The agent management feature uses the sudo command, so the sudo package must be installed in advance.

Agent Management Overview

The agents include a performance collection agent, a log collection agent, and a Windows event log collection agent.

  • The agent must be manually installed by the user on each monitoring target according to the user’s requirements.

Manage Agents

Managing Performance Agents

To install and manage the agent, follow these steps.

  1. Cloud Monitoring Console > Performance Analysis Click the button. You will be taken to the Performance Analysis page.
  2. On the Performance Analysis page, select a monitoring target and click the View Details button. The Monitoring Target Details popup window opens.
  3. Monitoring Target Details In the popup window, click the Agent tab. It navigates to the Agent tab.
  1. Click the Performance button in the Agent tab.
  2. Click the Copy icon on the right side of the installation command to copy the command.
  3. Paste the copied command into the monitoring target resource.
  4. Execute the command copied to the monitoring target resource.
Reference
The command uses sudo, so the sudo package must be installed.
Itemdescription
InstallationDownload the script file required for agent installation and execute it.
StartExecute the agent start command.
StopExecute the agent stop command.
DeleteExecute the agent deletion command.
UpdateDownload the script file required for the agent update and execute it.
Table. Managing Performance Agents
Reference

To check the agent service status, use the method below.

  • linux: $ sudo systemctl status metricbeat
  • windows: Task Manager → service → metricbeat → Status(Running)

Managing Log Agents

To install and manage the agent, follow these steps.

  1. Click Cloud Monitoring Console > Performance Analysis. You will be taken to the Performance Analysis page.
  2. On the Performance Analysis page, select a monitoring target and click the View Details button. The Monitoring Target Details popup window opens.
  3. Monitoring Target Details Click the Agent tab in the popup window. It navigates to the Agent tab.
  1. Click the Log button.
  2. Click the Copy icon on the right side of the installation command to copy the command.
  3. Paste the copied command into the monitoring target resource.
  4. Execute the command copied to the monitoring target resource.
Reference
Since the command uses sudo, the sudo package must be installed.
ItemExplanation
InstallationDownload the script file required for agent installation and execute it.
StartExecute the agent start command.
StopExecute the agent stop command.
DeleteExecute the agent deletion command.
UpdateDownload the script file required for the agent update and execute it.
Table. Managing Log Agents
Reference

To check the agent service status, use the method below.

  • linux: $ sudo systemctl status filebeat
  • windows: Task Manager → service → filebeat → Status(Running)

To add a log for monitoring, select the log addition action, enter the log name and log path correctly, and then click the Generate Command button. Paste the generated command into the monitored resource and then execute it.

Managing Event Agents

To install and manage the agent, follow the steps below.

  1. Click Cloud Monitoring Console > Performance Analysis. You will be taken to the Performance Analysis page.
  2. On the Performance Analysis page, select a monitoring target and click the View Details button. The Monitoring Target Details popup window opens.
  3. Monitoring Target Details In the popup window, click the Agent tab. It navigates to the Agent tab.
  1. Click the Event button.
  2. Click the Copy icon on the right of the installation command to copy the command.
  3. Paste the copied command into the monitoring target resource.
  4. Execute the command copied to the monitoring target resource.
Reference
The event agent is a target for Windows instance provisioning.
ItemExplanation
InstallationDownload the script file required for agent installation and execute it.
StartExecute the agent start command.
StopExecute the agent stop command.
DeleteExecute the agent deletion command.
UpdateDownload and run the script file required for the agent update.
Table. Managing Event Agents
Reference

To check the agent service status, use the method below.

  • windows: Task Manager → service → winlogbeat → Status(Running)
Caution
Agent command provision is offered independently of the Instance status of the Virtual Server (Bare Metal Server).

2.7 - Appendix A. Service-specific Monitoring Targets

Compute type

Virtual Server

CategoryMonitoring targetCollection methodCollection interval
PerformanceOSAgent
Agentless
1m
logOSAgentWhen a log occurs
StatusOSAgentless1m
Table. Virtual Server Monitoring Information
Reference
If you change the Virtual Server’s server type, monitoring performance metric data may not be collected correctly for a short period. Normal performance metrics will be collected in the next collection cycle (1 minute).

GPU Server

CategoryMonitoring targetCollection methodCollection interval
PerformanceOSAgent
Agentless
1m
logOSAgentWhen a log occurs
statusOSAgentless1m
Table. GPU Server Monitoring Information

Bare Metal Server

CategoryMonitoring targetCollection methodCollection interval
PerformanceOSAgent1m
logOSAgentWhen a log occurs
statusOSN/A-
Table. Bare Metal Server Monitoring Information

Multi-node GPU Cluster [Cluster Fabric]

CategoryMonitoring targetCollection methodCollection interval
PerformanceOSAgent1m
logOSAgentWhen a log occurs
statusOSN/A-
Table. Multi-node GPU Cluster [Cluster Fabric] Monitoring Information

Multi-node GPU Cluster [Node]

CategoryMonitoring targetCollection methodCollection interval
PerformanceOSAgent1m
logOSAgentWhen a log occurs
StatusOSN/A-
Table. Multi-node GPU Cluster [Node] Monitoring Information

Storage type

The monitoring targets, collection methods, and collection intervals are the same for all storage-type services.

  • File Storage
  • Object Storage
  • Block Storage(BM)
  • Block Storage(VM)
CategoryMonitoring targetCollection methodCollection interval
PerformanceStorageAgentless1m
logStorageN/A-
statusStorageAgentless1m
Table. Storage type monitoring information

Database type

The monitoring targets, collection methods, and collection intervals are the same for all Database-type services.

  • PostgreSQL(DBaaS)
  • MariaDB(DBaaS)
  • MySQL(DBaaS)
  • Microsoft SQL Server
  • EPAS
  • CacheStore(DBaaS)
    • Redis
    • Valkey
CategoryMonitoring targetCollection methodCollection interval
PerformanceDatabase Process, OSAgent1m
logDatabase Process, OSAgentWhen a log occurs
statusDatabase ProcessAgent1m
OSAgentless1m
Table. Database type monitoring information

Data Analytics type

CategoryMonitoring targetCollection methodCollection interval
PerformanceData Analytics Process, OSAgent1m
logData Analytics Process, OSAgentWhen a log occurs
statusData Analytics ProcessAgent1m
OSAgentless1m
Table. Data Analytics type monitoring information

Container type

Kubernetes Engine

CategoryMonitoring targetCollection methodCollection interval
PerformanceCluster, Namespace, Node, ReplicaSet, Deployment, StatefulSet, DaemonSet, Job, CronJob, PodAgentless5m
logCluster, Namespace, Node, ReplicaSet, Deployment, StatefulSet, DaemonSet, Job, CronJob, PodAgentlessWhen a log occurs
statusCluster, Namespace, Node, ReplicaSet, Deployment, StatefulSet, DaemonSet, Job, CronJob, PodAgentless5m
Table. Kubernetes Engine Monitoring Information

Container Registry

CategoryMonitoring targetCollection methodCollection interval
PerformanceContainer RegistryAgentless5m
logContainer RegistryAgentlessWhen a log occurs
statusContainer RegistryAgentless5m
Table. Container Registry monitoring information

Networking type

VPC

CategoryMonitoring targetCollection methodCollection interval
PerformanceInternet GatewayAgentless5m
logInternet GatewayN/A-
statusInternet GatewayN/A-
Table. Internet Gateway Monitoring Information
Caution
Performance monitoring is only possible when an Internet Gateway has been created.

Load Balancer(OLD)

Load Balancer(OLD)

CategoryMonitoring targetCollection methodCollection interval
PerformanceLoad BalencerAgentless5m
logLoad BalencerN/A-
statusLoad BalencerAgentless5m
Table. Load Balancer Monitoring Information

Load Balancer Listener(OLD)

CategoryMonitoring targetCollection methodCollection interval
PerformanceLoad Balencer ListenerAgentless5m
logLoad Balencer ListenerN/A-
statusLoad Balencer ListenerAgentless5m
Table. Load Balancer Listener Monitoring Information

Load Balancer

Load Balancer

CategoryMonitoring targetCollection methodCollection interval
PerformanceLoad BalencerAgentless5m
logLoad BalencerN/A-
statusLoad BalencerAgentless5m
Table. Load Balancer Monitoring Information

Load Balancer Listener

CategoryMonitoring targetCollection methodCollection interval
PerformanceLoad Balencer ListenerAgentless5m
logLoad Balencer ListenerN/A-
statusLoad Balencer ListenerAgentless5m
Table. Load Balancer Listener Monitoring Information

Load Balancer Server Group

CategoryMonitoring targetCollection methodCollection interval
PerformanceLoad Balencer Server GroupAgentless5m
logLoad Balencer Server GroupN/A-
statusLoad Balencer Server GroupAgentless5m
Table. Load Balancer Server Group Monitoring Information

Direct Connect

CategoryMonitoring targetCollection methodCollection interval
PerformanceDirect ConnectAgentless5m
logDirect ConnectN/A-
statusDirect ConnectN/A-
Table. Direct Connect Monitoring Information

Cloud WAN

CategoryMonitoring targetCollection methodCollection interval
PerformanceCloud WANAgentless10m
logCloud WANN/A-
statusCloud WANAgentless10m
Table. Cloud WAN Monitoring Information

Global CDN

CategoryMonitoring targetCollection methodCollection interval
PerformanceGlobal CDNAgentless5m
logGlobal CDNN/A-
statusGlobal CDNAgentless5m
Table. Global CDN Monitoring Information

2.8 - Appendix B. Service-specific Performance Metrics

Compute type

Virtual Server

Agentless (basic metrics)

Performance Item Group NamePerformance item namecollection unitCollection intervaldescription
CPUCPU Usage/Core [Basic]%1mPercentage of CPU time used, excluding Idle and IOWait states (normalized by the number of cores; 100% when all four cores are fully utilized)
CPUCPU Cores [Basic]cnt1mNumber of virtual processor cores allocated to the virtual machine
MemoryMemory Total [Basic]bytes1mMemory capacity available for use in the domain
MemoryMemory Used [Basic]bytes1mCurrent memory usage
MemoryMemory Swap In [Basic]bytes1mSwap In memory in bytes
MemoryMemory Swap Out [Basic]bytes1mSwap Out memory in bytes
MemoryMemory Free [Bytes]bytes1mUnused memory capacity in the system
MemoryMemory Usage [Basic]%1mCurrent memory usage rate
DiskDisk Read Bytes [Basic]bytes1mRead byte count
DiskDisk Read Requests [Basic]cnt1mRead request count
DiskDisk Write Bytes [Basic]bytes1mWrite byte count
DiskDisk Write Requests [Basic]cnt1mNumber of write requests
StateInstance State [Basic]enum1mVM status
NetworkNetwork In Bytes [Basic]bytes1mReceived bytes
NetworkNetwork In Dropped [Basic]cnt1mIncoming packet drop
NetworkNetwork In Errors [Basic]cnt1mReceive error
NetworkNetwork In Packets [Basic]cnt1mReceived packet
NetworkNetwork Out Bytes [Basic]bytes1mTransmit bytes
NetworkNetwork Out Dropped [Basic]cnt1mTransmit packet drop
NetworkNetwork Out Errors [Basic]cnt1mTransmission error
NetworkNetwork Out Packets [Basic]cnt1mTransmit packet
NetworkNetwork In Bytes [Delta Basic]bytes1mReceived bytes (delta value)
NetworkNetwork In Dropped [Delta Basic]cnt1mReceived packet drop (delta value)
NetworkNetwork In Errors [Delta Basic]cnt1mReceive error (delta value)
NetworkNetwork In Packets [Delta Basic]cnt1mReceived packet (delta value)
NetworkNetwork Out Bytes [Delta Basic]bytes1mTransmitted bytes (delta value)
NetworkNetwork Out Dropped [Delta Basic]cnt1mTransmit packet drop (delta value)
NetworkNetwork Out Errors [Delta Basic]cnt1mTransmission error (delta value)
NetworkNetwork Out Packets [Delta Basic]cnt1mTransmitted packet (delta value)
Table. Virtual Server (Agentless) Performance Metrics
Reference
  • For Windows OS, you must install the monitoring performance Agent to provide memory performance metrics.

Agent (Detailed Metrics)

Performance Item Group NamePerformance item namecollection unitCollection intervaldescription
CPUCore Usage [IO Wait]%1mRatio of CPU time spent in wait state (disk wait)
CPUCore Usage [System]%1mProportion of CPU time spent in kernel space
CPUCore Usage [User]%1mProportion of CPU time spent in user space
CPUCPU Corescnt1mThe 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 the maximum value is 100%* of a core.
CPUCPU Usage [Active]%1mPercentage of CPU time used excluding Idle and IOWait states (when all 4 cores are used at 100%: 400%)
CPUCPU Usage [Idle]%1mIt is the proportion of CPU time spent in idle state.
CPUCPU Usage [IO Wait]%1mIt is the proportion of CPU time spent in a waiting state (disk wait).
CPUCPU Usage [System]%1mPercentage of CPU time used by the kernel (when all 4 cores are used at 100%: 400%)
CPUCPU Usage [User]%1mPercentage of CPU time used in user space. (If all four cores are used at 100%, it is 400%)
CPUCPU Usage/Core [Active]%1mPercentage of CPU time used excluding Idle and IOWait states (value normalized by the number of cores; 100% when all four cores are fully utilized)
CPUCPU Usage/Core [Idle]%1mIt is the proportion of CPU time spent in idle state.
CPUCPU Usage/Core [IO Wait]%1mIt is the proportion of CPU time spent in a waiting state (disk wait).
CPUCPU Usage/Core [System]%1mPercentage of CPU time used by the kernel (value normalized by the number of cores; 100% when all four cores are fully utilized)
CPUCPU Usage/Core [User]%1mPercentage of CPU time used in user space. (Value normalized by the number of cores; using all four cores at 100% equals 100%)
DiskDisk CPU Usage [IO Request]%1mIt 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.
DiskDisk Queue Size [Avg]num1mThe average queue length of requests executed on the device.
DiskDisk Read Bytesbytes1mThe number of bytes per second read from the device.
DiskDisk Read Bytes [Delta Avg]bytes1mAverage of system.diskio.read.bytes_delta for individual disks
DiskDisk Read Bytes [Delta Max]bytes1mMaximum system.diskio.read.bytes_delta of individual disks
DiskDisk Read Bytes [Delta Min]bytes1mMinimum of system.diskio.read.bytes_delta for individual disks
DiskDisk Read Bytes [Delta Sum]bytes1mSum of system.diskio.read.bytes_delta for individual disks
DiskDisk Read Bytes [Delta]bytes1mDelta of the system.diskio.read.bytes value for each disk
DiskDisk Read Bytes [Success]bytes1mTotal number of bytes successfully read. On Linux, assuming a sector size of 512, it is the number of sectors read multiplied by 512.
DiskDisk Read Requestscnt1mNumber of read requests to the disk device per second
DiskDisk Read Requests [Delta Avg]cnt1mAverage of system.diskio.read.count_delta for individual disks
DiskDisk Read Requests [Delta Max]cnt1mMaximum of system.diskio.read.count_delta for individual disks
DiskDisk Read Requests [Delta Min]cnt1mMinimum of system.diskio.read.count_delta for individual disks
DiskDisk Read Requests [Delta Sum]cnt1mSum of system.diskio.read.count_delta for individual disks
DiskDisk Read Requests [Success Delta]cnt1mDelta of system.diskio.read.count for each disk
DiskDisk Read Requests [Success]cnt1mTotal number of reads successfully completed
DiskDisk Request Size [Avg]num1mAverage size of requests executed on the device (unit: sectors).
DiskDisk Service Time [Avg]ms1mAverage service time (ms) of input requests executed on the device.
DiskDisk Wait Time [Avg]ms1mAverage time taken for requests executed on the supported device.
DiskDisk Wait Time [Read]ms1mAverage disk wait time
DiskDisk Wait Time [Write]ms1mDisk average wait time
DiskDisk Write Bytes [Delta Avg]bytes1mAverage of system.diskio.write.bytes_delta for individual disks
DiskDisk Write Bytes [Delta Max]bytes1mMaximum system.diskio.write.bytes_delta of individual disks
DiskDisk Write Bytes [Delta Min]bytes1mMinimum of system.diskio.write.bytes_delta for individual disks
DiskDisk Write Bytes [Delta Sum]bytes1mSum of system.diskio.write.bytes_delta for individual disks
DiskDisk Write Bytes [Delta]bytes1mDelta of the system.diskio.write.bytes value for each individual disk
DiskDisk Write Bytes [Success]bytes1mTotal number of bytes successfully written. On Linux, assuming a sector size of 512, it is the number of sectors written multiplied by 512.
DiskDisk Write Requestscnt1mNumber of write requests to the disk device per second
DiskDisk Write Requests [Delta Avg]cnt1mAverage of system.diskio.write.count_delta for individual disks
DiskDisk Write Requests [Delta Max]cnt1mMaximum of system.diskio.write.count_delta for individual disks
DiskDisk Write Requests [Delta Min]cnt1mMinimum of system.diskio.write.count_delta for individual disks
DiskDisk Write Requests [Delta Sum]cnt1mSum of system.diskio.write.count_delta for individual disks
DiskDisk Write Requests [Success Delta]cnt1mDelta of system.diskio.write.count for each disk
DiskDisk Write Requests [Success]cnt1mTotal number of successful writes
DiskDisk Writes Bytesbytes1mThe number of bytes per second written to the device.
FileSystemFilesystem Hang Checkstate1mfilesystem (local/NFS) hang check (normal:1, abnormal:0)
FileSystemFilesystem Nodescnt1mTotal number of file nodes in the file system.
FileSystemFilesystem Nodes [Free]cnt1mIt is the total number of available file nodes in the file system.
FileSystemFilesystem Size [Available]bytes1mDisk space (bytes) that an unauthorized user can use.
FileSystemFilesystem Size [Free]bytes1mAvailable disk space (bytes)
FileSystemFilesystem Size [Total]bytes1mTotal disk space (bytes)
FileSystemFilesystem Usage%1mUsed disk space percentage
FileSystemFilesystem Usage [Avg]%1mAverage of individual filesystem.used.pct
FileSystemFilesystem Usage [Inode]%1minode usage
FileSystemFilesystem Usage [Max]%1mmax among individual filesystem.used.pct
FileSystemFilesystem Usage [Min]%1mminimum among individual filesystem.used.pct
FileSystemFilesystem Usage [Total]%1m-
FileSystemFilesystem Usedbytes1mUsed disk space (bytes)
FileSystemFilesystem Used [Inode]bytes1minode usage
MemoryMemory Freebytes1mTotal amount of available memory (bytes). Memory used by system cache and buffers is not included (see system.memory.actual.free).
MemoryMemory Free [Actual]bytes1mActual usable memory (bytes). The calculation method varies by OS; on Linux, it uses MemAvailable from /proc/ meminfo, or if meminfo cannot be used, it calculates from available memory plus cache and buffers. On OSX, it is the sum of usable memory and inactive memory. On Windows, it is a value such as system.memory.free.
MemoryMemory Free [Swap]bytes1mAvailable swap memory.
MemoryMemory Totalbytes1mtotal memory
MemoryMemory Total [Swap]bytes1mTotal swap memory.
MemoryMemory Usage%1mPercentage of used memory
MemoryMemory Usage [Actual]%1mPercentage of memory actually used
MemoryMemory Usage [Cache Swap]%1mcached swap usage
MemoryMemory Usage [Swap]%1mPercentage of used swap memory
MemoryMemory Usedbytes1mused memory
MemoryMemory Used [Actual]bytes1mActual memory used (bytes). The value obtained by subtracting used memory from total memory. Available memory is calculated differently for each OS (see system.actual.free).
MemoryMemory Used [Swap]bytes1mUsed swap memory.
NetworkCollisionscnt1mNetwork collision
NetworkNetwork In Bytesbytes1mNumber of received bytes
NetworkNetwork In Bytes [Delta Avg]bytes1mAverage of system.network.in.bytes_delta for individual networks
NetworkNetwork In Bytes [Delta Max]bytes1mMaximum of system.network.in.bytes_delta for each network
NetworkNetwork In Bytes [Delta Min]bytes1mMinimum of system.network.in.bytes_delta for each network
NetworkNetwork In Bytes [Delta Sum]bytes1mSum of system.network.in.bytes_delta for each network
NetworkNetwork In Bytes [Delta]bytes1mDelta of received byte count
NetworkNetwork In Droppedcnt1mNumber of deleted packets among incoming packets
NetworkNetwork In Errorscnt1mNumber of errors during reception
NetworkNetwork In Packetscnt1mNumber of received packets
NetworkNetwork In Packets [Delta Avg]cnt1mAverage of system.network.in.packets_delta for each network
NetworkNetwork In Packets [Delta Max]cnt1mMaximum of system.network.in.packets_delta for each network
NetworkNetwork In Packets [Delta Min]cnt1mMinimum of system.network.in.packets_delta for each network
NetworkNetwork In Packets [Delta Sum]cnt1mSum of system.network.in.packets_delta for each network
NetworkNetwork In Packets [Delta]cnt1mDelta of received packet count
NetworkNetwork Out Bytesbytes1mNumber of transmitted bytes
NetworkNetwork Out Bytes [Delta Avg]bytes1mAverage of system.network.out.bytes_delta for each network
NetworkNetwork Out Bytes [Delta Max]bytes1mMaximum of system.network.out.bytes_delta for each network
NetworkNetwork Out Bytes [Delta Min]bytes1mMinimum system.network.out.bytes_delta for each network
NetworkNetwork Out Bytes [Delta Sum]bytes1mSum of system.network.out.bytes_delta of individual networks
NetworkNetwork Out Bytes [Delta]bytes1mDelta of transmitted byte count
NetworkNetwork Out Droppedcnt1mNumber of packets deleted among outgoing packets. This value is not reported by the operating system, so it is always 0 on Darwin and BSD.
NetworkNetwork Out Errorscnt1mNumber of errors during transmission
NetworkNetwork Out Packetscnt1mNumber of transmitted packets
NetworkNetwork Out Packets [Delta Avg]cnt1mAverage of system.network.out.packets_delta for each network
NetworkNetwork Out Packets [Delta Max]cnt1mMaximum of system.network.out.packets_delta for each network
NetworkNetwork Out Packets [Delta Min]cnt1mMinimum of system.network.out.packets_delta for each network
NetworkNetwork Out Packets [Delta Sum]cnt1mSum of system.network.out.packets_delta for each network
NetworkNetwork Out Packets [Delta]cnt1mDelta of transmitted packet count
NetworkOpen Connections [TCP]cnt1mAll open TCP connections
NetworkOpen Connections [UDP]cnt1mAll open UDP connections
NetworkPort Usage%1mUsage rate of connectable ports
NetworkSYN Sent Socketscnt1mNumber of sockets in SYN_SENT state (when connecting from local to remote)
ProcessKernel PID Maxcnt1mkernel.pid_max value
ProcessKernel Thread Maxcnt1mkernel.threads-max value
ProcessProcess CPU Usage%1mThe percentage 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.
ProcessProcess CPU Usage/Core%1mThe percentage of CPU time used by the process since the last event. Normalized by the number of cores, with a value between 0 and 100%.
ProcessProcess Memory Usage%1mProportion of main memory (RAM) occupied by the process
ProcessProcess Memory Usedbytes1mResident Set size. The amount of memory a process occupies in RAM. In Windows, it is the current working set size.
ProcessProcess PIDPID1mprocess pid
ProcessProcess PPIDPID1mPID of the parent process
ProcessProcesses [Dead]cnt1mNumber of dead processes
ProcessProcesses [Idle]cnt1mNumber of idle processes
ProcessProcesses [Running]cnt1mrunning processes count
ProcessProcesses [Sleeping]cnt1msleeping processes count
ProcessProcesses [Stopped]cnt1mstopped processes count
ProcessProcesses [Total]cnt1mTotal number of processes
ProcessProcesses [Unknown]cnt1mNumber of processes with an unknown or unsearchable status
ProcessProcesses [Zombie]cnt1mNumber of zombie processes
ProcessRunning Process Usage%1mprocess usage
ProcessRunning Processescnt1mrunning processes count
ProcessRunning Thread Usage%1mThread usage rate
ProcessRunning Threadscnt1mTotal number of threads running in running processes
SystemContext Switchescnt1mcontext switch count (per second)
SystemLoad/Core [1 min]cnt1mThe load over the last 1 minute divided by the number of cores
SystemLoad/Core [15 min]cnt1mThe load over the last 15 minutes divided by the number of cores
SystemLoad/Core [5 min]cnt1mThe load over the last 5 minutes divided by the number of cores
SystemMultipaths [Active]cnt1mExternal storage connection path state = active count
SystemMultipaths [Failed]cnt1mExternal storage connection path state = failed count
SystemMultipaths [Faulty]cnt1mExternal storage connection path state = faulty count
SystemNTP Offsetnum1mthe measured offset of the last sample (time difference between the NTP server and the local environment)
SystemRun Queue Lengthnum1mExecution queue length
SystemUptimems1mOS uptime (milliseconds).
WindowsContext Switchiescnt1mCPU context switch count (per second)
WindowsDisk Read Bytes [Sec]cnt1mBytes read per second on a Windows logical disk
WindowsDisk Read Time [Avg]sec1mAverage data read time (seconds)
WindowsDisk Transfer Time [Avg]sec1mDisk average wait time
WindowsDisk Usage%1mDisk usage
WindowsDisk Write Bytes [Sec]cnt1mNumber of bytes written in one second on a Windows logical disk
WindowsDisk Write Time [Avg]sec1mAverage data write time (seconds)
WindowsPagingfile Usage%1mPaging file usage
WindowsPool Used [Non Paged]bytes1mNonpaged Pool usage in kernel memory
WindowsPool Used [Paged]bytes1mPaged Pool usage in kernel memory
WindowsProcess [Running]cnt1mNumber of processes currently running
WindowsThreads [Running]cnt1mNumber of threads currently running
WindowsThreads [Waiting]cnt1mNumber of threads waiting for processor time
Table. Virtual Server (Agent) Performance Items

GPU Server

Agentless (basic metrics)

Performance Item Group NamePerformance item namecollection unitCollection intervaldescription
CPUCPU Usage/Core [Basic]%1mPercentage of CPU time used, excluding Idle and IOWait states (normalized by the number of cores; 100% when all four cores are fully utilized)
CPUCPU Cores [Basic]cnt1mNumber of virtual processor cores allocated to the virtual machine
MemoryMemory Total [Basic]bytes1mMemory capacity available in the domain
MemoryMemory Used [Basic]bytes1mThe amount of memory currently in use
MemoryMemory Swap In [Basic]bytes1mSwap In memory in bytes
MemoryMemory Swap Out [Basic]bytes1mSwap Out memory in bytes
MemoryMemory Free [Bytes]bytes1mUnused memory capacity in the system
MemoryMemory Usage [Basic]%1mCurrent memory usage rate
DiskDisk Read Bytes [Basic]bytes1mRead byte count
DiskDisk Read Requests [Basic]cnt1mRead request count
DiskDisk Write Bytes [Basic]bytes1mWrite byte count
DiskDisk Write Requests [Basic]cnt1mNumber of write requests
StateInstance State [Basic]enum1mVM status
NetworkNetwork In Bytes [Basic]bytes1mReceived bytes
NetworkNetwork In Dropped [Basic]cnt1mIncoming packet drop
NetworkNetwork In Errors [Basic]cnt1mReceive error
NetworkNetwork In Packets [Basic]cnt1mReceived packet
NetworkNetwork Out Bytes [Basic]bytes1mTransmit bytes
NetworkNetwork Out Dropped [Basic]cnt1mTransmit packet drop
NetworkNetwork Out Errors [Basic]cnt1mTransmission error
NetworkNetwork Out Packets [Basic]cnt1mtransmitted packet
NetworkNetwork In Bytes [Delta Basic]bytes1mReceived bytes (delta value)
NetworkNetwork In Dropped [Delta Basic]cnt1mReceived packet drop (delta value)
NetworkNetwork In Errors [Delta Basic]cnt1mReceive error (delta value)
NetworkNetwork In Packets [Delta Basic]cnt1mReceived packet (delta value)
NetworkNetwork Out Bytes [Delta Basic]bytes1mTransmitted bytes (delta value)
NetworkNetwork Out Dropped [Delta Basic]cnt1mTransmit packet drop (delta value)
NetworkNetwork Out Errors [Delta Basic]cnt1mTransmission error (delta value)
NetworkNetwork Out Packets [Delta Basic]cnt1mTransmitted packet (delta value)
Table. GPU Server (Agentless) Performance Metrics

Agent (Detailed Metrics)

Performance Item Group NamePerformance item namecollection unitCollection intervaldescription
GPUGPU Countcnt1mGPU count
GPUGPU Memory Usage%1mMemory usage
GPUGPU Memory Usedbytes1mMemory usage
GPUGPU Temperature1mGPU temperature
GPUGPU Usage%1mTotal GPU utilization sum (800% when all 8 GPUs are used at 100%)
GPUGPU Usage [Avg]%1mOverall average GPU utilization (%)
GPUGPU Power CapW1mMaximum power capacity of the GPU
GPUGPU Power UsageW1mCurrent GPU power usage
GPUGPU Memory Usage [Avg]%1mGPU Memory Uti. AVG
GPUGPU Count in usecnt1mNumber of GPUs currently utilized by jobs on the node
GPUExecution State for nvidia-smistate1mResult of running the nvidia-smi command
CPUCore Usage [IO Wait]%1mRatio of CPU time spent in wait state (disk wait)
CPUCore Usage [System]%1mProportion of CPU time spent in kernel space
CPUCore Usage [User]%1mProportion of CPU time spent in user space
CPUCPU Corescnt1mThe 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 the maximum value is 100%* of a core.
CPUCPU Usage [Active]%1mPercentage of CPU time used excluding Idle and IOWait states (when all four cores are used at 100%: 400%)
CPUCPU Usage [Idle]%1mIt is the proportion of CPU time spent in idle state.
CPUCPU Usage [IO Wait]%1mIt is the proportion of CPU time spent in a waiting state (disk wait).
CPUCPU Usage [System]%1mCPU time usage percentage in the kernel (when all 4 cores are used at 100%: 400%)
CPUCPU Usage [User]%1mPercentage of CPU time used in user space. (If all 4 cores are used at 100%, it is 400%)
CPUCPU Usage/Core [Active]%1mPercentage of CPU time used excluding Idle and IOWait states (value normalized by the number of cores; 100% when all four cores are fully utilized)
CPUCPU Usage/Core [Idle]%1mIt is the proportion of CPU time spent in idle state.
CPUCPU Usage/Core [IO Wait]%1mIt is the proportion of CPU time spent in a waiting state (disk wait).
CPUCPU Usage/Core [System]%1mPercentage of CPU time used by the kernel (value normalized by the number of cores; 100% when all four cores are utilized at 100%)
CPUCPU Usage/Core [User]%1mPercentage of CPU time used in user space. (Value normalized by the number of cores; using all four cores at 100% equals 100%)
DiskDisk CPU Usage [IO Request]%1mThe proportion of CPU time during which I/O requests to the device were executed (device bandwidth utilization). If this value approaches 100%, the device becomes saturated.
DiskDisk Queue Size [Avg]num1mThe average queue length of requests executed for the device.
DiskDisk Read Bytesbytes1mThe number of bytes read per second from the device.
DiskDisk Read Bytes [Delta Avg]bytes1mAverage of system.diskio.read.bytes_delta for individual disks
DiskDisk Read Bytes [Delta Max]bytes1mMaximum system.diskio.read.bytes_delta of individual disks
DiskDisk Read Bytes [Delta Min]bytes1mMinimum of system.diskio.read.bytes_delta for individual disks
DiskDisk Read Bytes [Delta Sum]bytes1mSum of system.diskio.read.bytes_delta for individual disks
DiskDisk Read Bytes [Delta]bytes1mDelta of the system.diskio.read.bytes value for each disk
DiskDisk Read Bytes [Success]bytes1mTotal number of bytes successfully read. On Linux, assuming a sector size of 512, it is the number of sectors read multiplied by 512.
DiskDisk Read Requestscnt1mNumber of read requests to the disk device per second
DiskDisk Read Requests [Delta Avg]cnt1mAverage of system.diskio.read.count_delta for individual disks
DiskDisk Read Requests [Delta Max]cnt1mMaximum system.diskio.read.count_delta of individual disks
DiskDisk Read Requests [Delta Min]cnt1mMinimum of system.diskio.read.count_delta for individual disks
DiskDisk Read Requests [Delta Sum]cnt1mSum of system.diskio.read.count_delta for individual disks
DiskDisk Read Requests [Success Delta]cnt1mDelta of system.diskio.read.count for each disk
DiskDisk Read Requests [Success]cnt1mTotal number of successful reads
DiskDisk Request Size [Avg]num1mAverage size of requests executed on the device (unit: sectors).
DiskDisk Service Time [Avg]ms1mAverage service time (ms) of input requests executed on the device.
DiskDisk Wait Time [Avg]ms1mAverage time taken for requests executed on the supported device.
DiskDisk Wait Time [Read]ms1mAverage disk wait time
DiskDisk Wait Time [Write]ms1mAverage disk wait time
DiskDisk Write Bytes [Delta Avg]bytes1mAverage of system.diskio.write.bytes_delta for individual disks
DiskDisk Write Bytes [Delta Max]bytes1mMaximum system.diskio.write.bytes_delta of individual disks
DiskDisk Write Bytes [Delta Min]bytes1mMinimum of system.diskio.write.bytes_delta for individual disks
DiskDisk Write Bytes [Delta Sum]bytes1mSum of system.diskio.write.bytes_delta for individual disks
DiskDisk Write Bytes [Delta]bytes1mDelta of the system.diskio.write.bytes value for each disk
DiskDisk Write Bytes [Success]bytes1mTotal number of bytes successfully written. On Linux, assuming a sector size of 512, it is the number of sectors written multiplied by 512.
DiskDisk Write Requestscnt1mNumber of write requests to the disk device per second
DiskDisk Write Requests [Delta Avg]cnt1mAverage of system.diskio.write.count_delta for individual disks
DiskDisk Write Requests [Delta Max]cnt1mMaximum of system.diskio.write.count_delta for individual disks
DiskDisk Write Requests [Delta Min]cnt1mMinimum of system.diskio.write.count_delta for individual disks
DiskDisk Write Requests [Delta Sum]cnt1mSum of system.diskio.write.count_delta for individual disks
DiskDisk Write Requests [Success Delta]cnt1mDelta of system.diskio.write.count for each disk
DiskDisk Write Requests [Success]cnt1mTotal number of successful writes
DiskDisk Writes Bytesbytes1mThe number of bytes per second written to the device.
FileSystemFilesystem Hang Checkstate1mfilesystem(local/NFS) hang check (normal:1, abnormal:0)
FileSystemFilesystem Nodescnt1mTotal number of file nodes in the file system.
FileSystemFilesystem Nodes [Free]cnt1mTotal number of available file nodes in the file system.
FileSystemFilesystem Size [Available]bytes1mDisk space (bytes) that an unauthorized user can use.
FileSystemFilesystem Size [Free]bytes1mAvailable disk space (bytes)
FileSystemFilesystem Size [Total]bytes1mTotal disk space (bytes)
FileSystemFilesystem Usage%1mUsed disk space percentage
FileSystemFilesystem Usage [Avg]%1mAverage of individual filesystem.used.pct
FileSystemFilesystem Usage [Inode]%1minode usage
FileSystemFilesystem Usage [Max]%1mmax among individual filesystem.used.pct
FileSystemFilesystem Usage [Min]%1mminimum among individual filesystem.used.pct
FileSystemFilesystem Usage [Total]%1m-
FileSystemFilesystem Usedbytes1mUsed disk space (bytes)
FileSystemFilesystem Used [Inode]bytes1minode usage
MemoryMemory Freebytes1mTotal amount of available memory (bytes). Does not include memory used by system cache and buffers (see system.memory.actual.free).
MemoryMemory Free [Actual]bytes1mActual usable memory (bytes). The calculation method varies by OS; on Linux, it is MemAvailable from /proc/meminfo, or if meminfo is unavailable, it is calculated from available memory plus cache and buffers. On macOS, it is the sum of usable memory and inactive memory. On Windows, it is a value such as system.memory.free.
MemoryMemory Free [Swap]bytes1mAvailable swap memory.
MemoryMemory Totalbytes1mtotal memory
MemoryMemory Total [Swap]bytes1mTotal swap memory.
MemoryMemory Usage%1mPercentage of used memory
MemoryMemory Usage [Actual]%1mPercentage of memory actually used
MemoryMemory Usage [Cache Swap]%1mCached swap usage
MemoryMemory Usage [Swap]%1mPercentage of used swap memory
MemoryMemory Usedbytes1mused memory
MemoryMemory Used [Actual]bytes1mActual memory used (bytes). The value obtained by subtracting used memory from total memory. Available memory is calculated differently for each OS (see system.actual.free).
MemoryMemory Used [Swap]bytes1mUsed swap memory.
NetworkCollisionscnt1mNetwork collision
NetworkNetwork In Bytesbytes1mNumber of received bytes
NetworkNetwork In Bytes [Delta Avg]bytes1mAverage of system.network.in.bytes_delta for individual networks
NetworkNetwork In Bytes [Delta Max]bytes1mMaximum of system.network.in.bytes_delta for each network
NetworkNetwork In Bytes [Delta Min]bytes1mMinimum of system.network.in.bytes_delta for each network
NetworkNetwork In Bytes [Delta Sum]bytes1mSum of system.network.in.bytes_delta for each network
NetworkNetwork In Bytes [Delta]bytes1mDelta of received byte count
NetworkNetwork In Droppedcnt1mNumber of deleted packets among incoming packets
NetworkNetwork In Errorscnt1mNumber of errors during reception
NetworkNetwork In Packetscnt1mNumber of received packets
NetworkNetwork In Packets [Delta Avg]cnt1mAverage of system.network.in.packets_delta for each network
NetworkNetwork In Packets [Delta Max]cnt1mMaximum of system.network.in.packets_delta for each network
NetworkNetwork In Packets [Delta Min]cnt1mMinimum of system.network.in.packets_delta for each network
NetworkNetwork In Packets [Delta Sum]cnt1mSum of system.network.in.packets_delta for each network
NetworkNetwork In Packets [Delta]cnt1mDelta of received packet count
NetworkNetwork Out Bytesbytes1mNumber of transmitted bytes
NetworkNetwork Out Bytes [Delta Avg]bytes1mAverage of system.network.out.bytes_delta for each network
NetworkNetwork Out Bytes [Delta Max]bytes1mMaximum of system.network.out.bytes_delta for each network
NetworkNetwork Out Bytes [Delta Min]bytes1mMinimum of system.network.out.bytes_delta for each network
NetworkNetwork Out Bytes [Delta Sum]bytes1mSum of system.network.out.bytes_delta for individual networks
NetworkNetwork Out Bytes [Delta]bytes1mDelta of transmitted byte count
NetworkNetwork Out Droppedcnt1mNumber of packets deleted among outgoing packets. This value is not reported by the operating system, so it is always 0 on Darwin and BSD.
NetworkNetwork Out Errorscnt1mNumber of errors during transmission
NetworkNetwork Out Packetscnt1mNumber of transmitted packets
NetworkNetwork Out Packets [Delta Avg]cnt1mAverage of system.network.out.packets_delta for each network
NetworkNetwork Out Packets [Delta Max]cnt1mMaximum of system.network.out.packets_delta for each network
NetworkNetwork Out Packets [Delta Min]cnt1mMinimum of system.network.out.packets_delta for each network
NetworkNetwork Out Packets [Delta Sum]cnt1mSum of system.network.out.packets_delta for each network
NetworkNetwork Out Packets [Delta]cnt1mDelta of transmitted packet count
NetworkOpen Connections [TCP]cnt1mAll open TCP connections
NetworkOpen Connections [UDP]cnt1mAll open UDP connections
NetworkPort Usage%1mConnectable port utilization
NetworkSYN Sent Socketscnt1mNumber of sockets in SYN_SENT state (when connecting from local to remote)
ProcessKernel PID Maxcnt1mkernel.pid_max value
ProcessKernel Thread Maxcnt1mkernel.threads-max value
ProcessProcess CPU Usage%1mThe percentage 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.
ProcessProcess CPU Usage/Core%1mThe percentage of CPU time used by the process since the last event. Normalized by the number of cores, with a value between 0 and 100%.
ProcessProcess Memory Usage%1mThe proportion of main memory (RAM) occupied by the process
ProcessProcess Memory Usedbytes1mResident Set size. The amount of memory a process occupies in RAM. In Windows, it is the current working set size.
ProcessProcess PIDPID1mprocess pid
ProcessProcess PPIDPID1mParent process PID
ProcessProcesses [Dead]cnt1mNumber of dead processes
ProcessProcesses [Idle]cnt1mNumber of idle processes
ProcessProcesses [Running]cnt1mrunning processes count
ProcessProcesses [Sleeping]cnt1msleeping processes count
ProcessProcesses [Stopped]cnt1mstopped processes count
ProcessProcesses [Total]cnt1mTotal number of processes
ProcessProcesses [Unknown]cnt1mNumber of processes with an unknown or unsearchable status
ProcessProcesses [Zombie]cnt1mNumber of zombie processes
ProcessRunning Process Usage%1mprocess usage rate
ProcessRunning Processescnt1mrunning processes count
ProcessRunning Thread Usage%1mThread usage rate
ProcessRunning Threadscnt1mTotal number of threads running in running processes
SystemContext Switchescnt1mcontext switch count (per second)
SystemLoad/Core [1 min]cnt1mThe load over the last 1 minute divided by the number of cores
SystemLoad/Core [15 min]cnt1mThe load over the last 15 minutes divided by the number of cores
SystemLoad/Core [5 min]cnt1mThe load over the last 5 minutes divided by the number of cores
SystemMultipaths [Active]cnt1mExternal storage connection path state = active count
SystemMultipaths [Failed]cnt1mExternal storage connection path state = failed count
SystemMultipaths [Faulty]cnt1mExternal storage connection path state = faulty count
SystemNTP Offsetnum1mthe measured offset of the last sample (time difference between the NTP server and the local environment)
SystemRun Queue Lengthnum1mExecution queue length
SystemUptimems1mOS uptime (uptime). (milliseconds)
WindowsContext Switchiescnt1mCPU context switch count (per second)
WindowsDisk Read Bytes [Sec]cnt1mNumber of bytes read in one second from a Windows logical disk
WindowsDisk Read Time [Avg]sec1mAverage data read time (seconds)
WindowsDisk Transfer Time [Avg]sec1mDisk average wait time
WindowsDisk Usage%1mDisk usage
WindowsDisk Write Bytes [Sec]cnt1mBytes written per second on a Windows logical disk
WindowsDisk Write Time [Avg]sec1mAverage data write time (seconds)
WindowsPagingfile Usage%1mPaging file usage
WindowsPool Used [Non Paged]bytes1mNonpaged Pool usage in kernel memory
WindowsPool Used [Paged]bytes1mPaged Pool usage in kernel memory
WindowsProcess [Running]cnt1mNumber of processes currently running
WindowsThreads [Running]cnt1mNumber of threads currently running
WindowsThreads [Waiting]cnt1mNumber of threads waiting for processor time
Table. GPU Server (Agent) Performance Metrics

Bare Metal Server

Agent (detailed metrics)

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
CPUCore Usage [IO Wait]%1mProportion of CPU time spent waiting (disk wait)
CPUCore Usage [System]%1mProportion of CPU time spent in kernel space
CPUCore Usage [User]%1mProportion of CPU time spent in user space
CPUCPU Corescnt1mThe 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.
CPUCPU Usage [Active]%1mPercentage of CPU time used excluding Idle and IOWait states (when all 4 cores are used at 100%: 400%)
CPUCPU Usage [Idle]%1mIt is the proportion of CPU time spent in idle state.
CPUCPU Usage [IO Wait]%1mIt is the proportion of CPU time spent in a waiting state (disk wait).
CPUCPU Usage [System]%1mPercentage of CPU time used by the kernel (when all 4 cores are used at 100%: 400%)
CPUCPU Usage [User]%1mPercentage of CPU time used in user space. (If all 4 cores are used at 100%, it is 400%)
CPUCPU Usage/Core [Active]%1mPercentage of CPU time used excluding Idle and IOWait states (value normalized by the number of cores; 100% when all four cores are fully utilized)
CPUCPU Usage/Core [Idle]%1mIt is the proportion of CPU time spent in idle state.
CPUCPU Usage/Core [IO Wait]%1mIt is the proportion of CPU time spent in a waiting state (disk wait).
CPUCPU Usage/Core [System]%1mPercentage of CPU time used by the kernel (value normalized by the number of cores; 100% when all 4 cores are fully utilized)
CPUCPU Usage/Core [User]%1mPercentage of CPU time used in user space. (Value normalized by the number of cores; using all four cores at 100% each equals 100%)
DiskDisk CPU Usage [IO Request]%1mThe proportion of CPU time during which I/O requests to the device were executed (device bandwidth utilization). If this value approaches 100%, the device becomes saturated.
DiskDisk Queue Size [Avg]num1mThe average queue length of requests executed for the device.
DiskDisk Read Bytesbytes1mThe number of bytes read per second from the device.
DiskDisk Read Bytes [Delta Avg]bytes1mAverage of system.diskio.read.bytes_delta for individual disks
DiskDisk Read Bytes [Delta Max]bytes1mMaximum system.diskio.read.bytes_delta of individual disks
DiskDisk Read Bytes [Delta Min]bytes1mMinimum of system.diskio.read.bytes_delta for individual disks
DiskDisk Read Bytes [Delta Sum]bytes1mSum of system.diskio.read.bytes_delta for individual disks
DiskDisk Read Bytes [Delta]bytes1mDelta of the system.diskio.read.bytes value for each disk
DiskDisk Read Bytes [Success]bytes1mTotal number of bytes successfully read. On Linux, the sector size is assumed to be 512, and the value is the number of sectors read multiplied by 512.
DiskDisk Read Requestscnt1mNumber of read requests to the disk device per second
DiskDisk Read Requests [Delta Avg]cnt1mAverage of the system.diskio.read.count_delta for individual disks
DiskDisk Read Requests [Delta Max]cnt1mMaximum system.diskio.read.count_delta of individual disks
DiskDisk Read Requests [Delta Min]cnt1mMinimum of system.diskio.read.count_delta for individual disks
DiskDisk Read Requests [Delta Sum]cnt1mSum of system.diskio.read.count_delta for individual disks
DiskDisk Read Requests [Success Delta]cnt1mDelta of system.diskio.read.count for each disk
DiskDisk Read Requests [Success]cnt1mTotal number of successful reads
DiskDisk Request Size [Avg]num1mIt is the average size of requests executed on the device (unit: sectors).
DiskDisk Service Time [Avg]ms1mAverage service time (ms) of input requests executed on the device.
DiskDisk Wait Time [Avg]ms1mAverage time taken for requests executed on the supported device.
DiskDisk Wait Time [Read]ms1mAverage disk wait time
DiskDisk Wait Time [Write]ms1mAverage disk wait time
DiskDisk Write Bytes [Delta Avg]bytes1mAverage of system.diskio.write.bytes_delta for individual disks
DiskDisk Write Bytes [Delta Max]bytes1mMaximum system.diskio.write.bytes_delta of individual disks
DiskDisk Write Bytes [Delta Min]bytes1mMinimum of system.diskio.write.bytes_delta for individual disks
DiskDisk Write Bytes [Delta Sum]bytes1mSum of system.diskio.write.bytes_delta for individual disks
DiskDisk Write Bytes [Delta]bytes1mDelta of the system.diskio.write.bytes value for each disk
DiskDisk Write Bytes [Success]bytes1mTotal number of bytes successfully written. On Linux, the sector size is assumed to be 512, and the value is the number of sectors written multiplied by 512.
DiskDisk Write Requestscnt1mNumber of write requests to the disk device per second
DiskDisk Write Requests [Delta Avg]cnt1mAverage of system.diskio.write.count_delta for individual disks
DiskDisk Write Requests [Delta Max]cnt1mMaximum of system.diskio.write.count_delta for individual disks
DiskDisk Write Requests [Delta Min]cnt1mMinimum of system.diskio.write.count_delta for individual disks
DiskDisk Write Requests [Delta Sum]cnt1mSum of system.diskio.write.count_delta for individual disks
DiskDisk Write Requests [Success Delta]cnt1mDelta of system.diskio.write.count for each disk
DiskDisk Write Requests [Success]cnt1mTotal number of successful writes
DiskDisk Writes Bytesbytes1mThe number of bytes per second written to the device.
FileSystemFilesystem Hang Checkstate1mfilesystem(local/NFS) hang check (normal:1, abnormal:0)
FileSystemFilesystem Nodescnt1mTotal number of file nodes in the file system.
FileSystemFilesystem Nodes [Free]cnt1mTotal number of available file nodes in the file system.
FileSystemFilesystem Size [Available]bytes1mDisk space (bytes) that an unauthorized user can use.
FileSystemFilesystem Size [Free]bytes1mAvailable disk space (bytes)
FileSystemFilesystem Size [Total]bytes1mTotal disk space (bytes)
FileSystemFilesystem Usage%1mUsed disk space percentage
FileSystemFilesystem Usage [Avg]%1mAverage of individual filesystem.used.pct
FileSystemFilesystem Usage [Inode]%1minode usage
FileSystemFilesystem Usage [Max]%1mmax among individual filesystem.used.pct
FileSystemFilesystem Usage [Min]%1mminimum among individual filesystem.used.pct
FileSystemFilesystem Usage [Total]%1m-
FileSystemFilesystem Usedbytes1mUsed disk space (bytes)
FileSystemFilesystem Used [Inode]bytes1minode usage
MemoryMemory Freebytes1mTotal amount of available memory (bytes). Does not include memory used by system cache and buffers (see system.memory.actual.free).
MemoryMemory Free [Actual]bytes1mActual 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 macOS, it is the sum of usable memory and inactive memory. On Windows, it is a value such as system.memory.free.
MemoryMemory Free [Swap]bytes1mAvailable swap memory.
MemoryMemory Totalbytes1mtotal memory
MemoryMemory Total [Swap]bytes1mTotal swap memory.
MemoryMemory Usage%1mPercentage of used memory
MemoryMemory Usage [Actual]%1mPercentage of memory actually used
MemoryMemory Usage [Cache Swap]%1mCached swap usage
MemoryMemory Usage [Swap]%1mPercentage of used swap memory
MemoryMemory Usedbytes1mused memory
MemoryMemory Used [Actual]bytes1mActual memory used (bytes). The value obtained by subtracting used memory from total memory. Available memory is calculated differently for each OS (see system.actual.free).
MemoryMemory Used [Swap]bytes1mUsed swap memory.
NetworkCollisionscnt1mNetwork collision
NetworkNetwork In Bytesbytes1mNumber of received bytes
NetworkNetwork In Bytes [Delta Avg]bytes1mAverage of system.network.in.bytes_delta for individual networks
NetworkNetwork In Bytes [Delta Max]bytes1mMaximum of system.network.in.bytes_delta for each network
NetworkNetwork In Bytes [Delta Min]bytes1mMinimum of system.network.in.bytes_delta for each network
NetworkNetwork In Bytes [Delta Sum]bytes1mSum of system.network.in.bytes_delta for each network
NetworkNetwork In Bytes [Delta]bytes1mDelta of received byte count
NetworkNetwork In Droppedcnt1mNumber of deleted packets among incoming packets
NetworkNetwork In Errorscnt1mNumber of errors during reception
NetworkNetwork In Packetscnt1mNumber of received packets
NetworkNetwork In Packets [Delta Avg]cnt1mAverage of system.network.in.packets_delta for each network
NetworkNetwork In Packets [Delta Max]cnt1mMaximum of system.network.in.packets_delta for each network
NetworkNetwork In Packets [Delta Min]cnt1mMinimum of system.network.in.packets_delta for each network
NetworkNetwork In Packets [Delta Sum]cnt1mSum of system.network.in.packets_delta for each network
NetworkNetwork In Packets [Delta]cnt1mDelta of received packet count
NetworkNetwork Out Bytesbytes1mNumber of transmitted bytes
NetworkNetwork Out Bytes [Delta Avg]bytes1mAverage of system.network.out.bytes_delta for each network
NetworkNetwork Out Bytes [Delta Max]bytes1mMaximum of system.network.out.bytes_delta for each network
NetworkNetwork Out Bytes [Delta Min]bytes1mMinimum of system.network.out.bytes_delta for each network
NetworkNetwork Out Bytes [Delta Sum]bytes1mSum of system.network.out.bytes_delta for individual networks
NetworkNetwork Out Bytes [Delta]bytes1mDelta of transmitted byte count
NetworkNetwork Out Droppedcnt1mNumber of deleted packets among outgoing packets. This value is not reported by the operating system, so it is always 0 on Darwin and BSD.
NetworkNetwork Out Errorscnt1mNumber of errors during transmission
NetworkNetwork Out Packetscnt1mNumber of transmitted packets
NetworkNetwork Out Packets [Delta Avg]cnt1mAverage of system.network.out.packets_delta for each network
NetworkNetwork Out Packets [Delta Max]cnt1mMaximum of system.network.out.packets_delta for each network
NetworkNetwork Out Packets [Delta Min]cnt1mMinimum of system.network.out.packets_delta for each network
NetworkNetwork Out Packets [Delta Sum]cnt1mSum of system.network.out.packets_delta for each network
NetworkNetwork Out Packets [Delta]cnt1mDelta of transmitted packet count
NetworkOpen Connections [TCP]cnt1mAll open TCP connections
NetworkOpen Connections [UDP]cnt1mAll open UDP connections
NetworkPort Usage%1mConnectable port utilization
NetworkSYN Sent Socketscnt1mNumber of sockets in SYN_SENT state (when connecting from local to remote)
ProcessKernel PID Maxcnt1mkernel.pid_max value
ProcessKernel Thread Maxcnt1mkernel.threads-max value
ProcessProcess CPU Usage%1mThe percentage 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.
ProcessProcess CPU Usage/Core%1mThe percentage of CPU time used by the process since the last event. Normalized by the number of cores, with a value between 0 and 100%.
ProcessProcess Memory Usage%1mThe proportion of main memory (RAM) occupied by the process
ProcessProcess Memory Usedbytes1mResident Set size. The amount of memory a process occupies in RAM. In Windows, it is the current working set size.
ProcessProcess PIDPID1mprocess pid
ProcessProcess PPIDPID1mParent process PID
ProcessProcesses [Dead]cnt1mNumber of dead processes
ProcessProcesses [Idle]cnt1mNumber of idle processes
ProcessProcesses [Running]cnt1mrunning processes count
ProcessProcesses [Sleeping]cnt1msleeping processes count
ProcessProcesses [Stopped]cnt1mstopped processes count
ProcessProcesses [Total]cnt1mTotal number of processes
ProcessProcesses [Unknown]cnt1mNumber of processes with an unknown or unsearchable status
ProcessProcesses [Zombie]cnt1mNumber of zombie processes
ProcessRunning Process Usage%1mprocess usage rate
ProcessRunning Processescnt1mrunning processes count
ProcessRunning Thread Usage%1mThread usage rate
ProcessRunning Threadscnt1mTotal number of threads running in running processes
SystemContext Switchescnt1mcontext switch count (per second)
SystemLoad/Core [1 min]cnt1mThe load over the last 1 minute divided by the number of cores
SystemLoad/Core [15 min]cnt1mThe load over the last 15 minutes divided by the number of cores
SystemLoad/Core [5 min]cnt1mThe load over the last 5 minutes divided by the number of cores
SystemMultipaths [Active]cnt1mExternal storage connection path state = active count
SystemMultipaths [Failed]cnt1mExternal storage connection path state = failed count
SystemMultipaths [Faulty]cnt1mExternal storage connection path state = faulty count
SystemNTP Offsetnum1mthe measured offset of the last sample (time difference between the NTP server and the local environment)
SystemRun Queue Lengthnum1mExecution queue length
SystemUptimems1mOS uptime (uptime). (milliseconds)
WindowsContext Switchiescnt1mCPU context switch count (per second)
WindowsDisk Read Bytes [Sec]cnt1mNumber of bytes read in one second from a Windows logical disk
WindowsDisk Read Time [Avg]sec1mAverage data read time (seconds)
WindowsDisk Transfer Time [Avg]sec1mDisk average wait time
WindowsDisk Usage%1mDisk usage
WindowsDisk Write Bytes [Sec]cnt1mBytes written per second on a Windows logical disk
WindowsDisk Write Time [Avg]sec1mAverage data write time (seconds)
WindowsPagingfile Usage%1mPaging file usage
WindowsPool Used [Non Paged]bytes1mNonpaged Pool usage in kernel memory
WindowsPool Used [Paged]bytes1mPaged Pool usage in kernel memory
WindowsProcess [Running]cnt1mNumber of processes currently running
WindowsThreads [Running]cnt1mNumber of threads currently running
WindowsThreads [Waiting]cnt1mNumber of threads waiting for processor time
Table. Bare Metal Server (Agent) Performance Items
reference
To monitor performance metrics of a Bare Metal Server, please install the Agent. Refer to Manage Agents for the Agent installation guide.

Multi-node GPU Cluster [Cluster Fabric]

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
Cluster GPUCluster GPU Countcnt1mCluster GPU Count Sum.
Sum of node GPU Count within the cluster: calculate the total GPU Count of each node within the same GPU cluster.
Cluster GPUCluster GPU Count In Usecnt1mNumber of GPUs being used by Jobs in the cluster
Number of GPUs used by Processes in the cluster: Parse the ‘Processes:’ section at the bottom of nvidia-smi output from nodes in the same GPU cluster and sum the number of GPUs held by processes
Cluster GPUCluster GPU Usage%1mGPU Utilization Average within the cluster.
GPU Utilization Average value for nodes within the cluster: calculate the average of each node’s GPU Utilization values among nodes in the same GPU cluster.
Cluster GPUCluster GPU Memory Usage [Avg]%1mGPU Memory Utilization Average within the Song cluster.
Cluster node Memory Utilization Average value: calculates the average of each node’s Memory Utilization values among nodes in the same GPU cluster.
Table. Multi-node GPU Cluster [Cluster Fabric] Performance Items

Multi-node GPU Cluster [Node]

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
GPUGPU Countcnt1mNumber of GPUs
GPUGPU Memory Usage%1mMemory usage
GPUGPU Memory UsedMB1mMemory usage
GPUGPU Temperature1mGPU temperature
GPUGPU Usage%1mUtilization
GPUGPU Usage [Avg]%1mOverall average GPU utilization (%)
GPUGPU Power CapW1mMaximum power capacity of the GPU
GPUGPU Power UsageW1mCurrent GPU power usage
GPUGPU Memory Usage [Avg]%1mGPU Memory Utilization Average
GPUGPU Count in usecnt1mNumber of GPUs in use by jobs on the node
GPUExecution State for nvidia-smistate1mResult of running the nvidia-smi command
Table. Multi-node GPU Cluster [Node] Performance Items
Reference
Refer to the Bare Metal Server’s performance items for OS performance metrics.

Storage type

File Storage

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
VolumeInstance Statestate1mfilestorage volume status
VolumeIOPS [Other]iops1miops (other)
VolumeIOPS [Read]iops1miops(read)
VolumeIOPS [Total]iops1miops(total)
VolumeIOPS [Write]iops1miops(write)
VolumeLatency Time [Other]usec1mLatency (Other)
VolumeLatency Time [Read]usec1mRead latency
VolumeLatency Time [Total]usec1mTotal latency
VolumeLatency Time [write]usec1mWrite latency
VolumeThroughput [Other]bytes/s1mThroughput (Other)
VolumeThroughput [Read]bytes/s1mThroughput (read)
VolumeThroughput [Total]bytes/s1mThroughput (total)
VolumeThroughput [Write]bytes/s1mThroughput (write)
VolumeVolume Totalbytes1mTotal byte count
VolumeVolume Usage%1mUsage rate
VolumeVolume Usedbytes1mUsage
Table. File Storage performance items

Object Storage

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
RequestRequests [Delete]cnt1mNumber of HTTP DELETE requests executed on objects in the bucket
RequestRequests [Download Avg]bytes1mDownload usage per bucket
RequestRequests [Get]cnt1mNumber of HTTP GET requests executed on objects in the bucket
RequestRequests [Head]cnt1mNumber of HTTP HEAD requests executed on objects in the bucket
RequestRequests [List]cnt1mNumber of LIST requests executed for objects in the bucket
RequestRequests [Post]cnt1mNumber of HTTP POST requests executed on objects in the bucket
RequestRequests [Put]cnt1mNumber of HTTP PUT requests executed on objects in the bucket
RequestRequests [Total]cnt1mTotal number of HTTP requests executed on the bucket
RequestRequests [Upload Avg]bytes1mUpload usage per bucket
UsageBucket Usedbytes1mAmount of data stored in the bucket (bytes)
UsageObjectscnt1mNumber of objects stored in the bucket
Table. Object Storage performance items

Block Storage(BM)

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
StateInstance Statestate1mBlockstorage volume status
VolumeIOPS [Total]iops1miops(total)
VolumeIOPS [Read]iops1miops(read)
VolumeIOPS [Write]iops1miops(write)
VolumeIOPS [Other]iops1miops (other)
VolumeLatency Time [Total]usec1mTotal latency
VolumeLatency Time [Read]usec1mRead latency
VolumeLatency Time [Write]usec1mWrite latency
VolumeLatency Time [Other]usec1mLatency (Other)
VolumeThroughput [Total]MB/s1mThroughput (total)
VolumeThroughput [Read]MB/s1mThroughput (read)
VolumeThroughput [Write]MB/s1mThroughput (write)
VolumeThroughput [Other]MB/s1mThroughput (Other)
VolumeVolume Bytesbytes1mTotal byte count
Table. Block Storage (BM) performance items

Block Storage(VM)

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
StateInstance Statestate1mBlockstorage volume status
VolumeIOPS [Read]iops1miops(read)
VolumeIOPS [Write]iops1miops(write)
VolumeLatency Time [Read]usec1mRead latency
VolumeLatency Time [Write]usec1mWrite latency
VolumeThroughput [Read]MB/s1mThroughput (read)
VolumeThroughput [Write]MB/s1mThroughput (write)
VolumeVolume Bytesbytes1mTotal byte count
Table. Block Storage (VM) performance items

Database type

PostgreSQL(DBaaS)

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
ActivelockActive Lockscnt1mNumber of activelocks
ActivelockActive Locks [Access Exclusive]cnt1maccessexclusive lock count
ActivelockActive Locks [Access Share]cnt1mNumber of accessshare locks
ActivelockActive Locks [Total]cnt1m-
ActivelockExclusive Lockscnt1mexclusive lock count
ActivelockRow Exclusive Lockscnt1mrow exclusive lock count
ActivelockRow Share Lockscnt1mrow share lock count
ActivelockShare Lockscnt1mshare lock count
ActivelockShare Row Exclusive Lockscnt1mNumber of sharerowexclusive locks
ActivelockShare Update Exclusive Lockscnt1mNumber of share update exclusive locks
ActiveSessionActive Sessionscnt1mNumber of active sessions
ActiveSessionActive Sessions [Total]cnt1m-
ActiveSessionIdle In Transaction Sessionscnt1mNumber of sessions in idle_in_transaction state
ActiveSessionIdle In Transaction Sessions [Total]cnt1m-
ActiveSessionIdle Sessionscnt1mNumber of idle sessions
ActiveSessionIdle Sessions [Total]cnt1m-
ActiveSessionWaiting Sessionscnt1mNumber of sessions in waiting state
ActiveSessionWaiting Sessions [Total]cnt1m-
ConnectionConnection Usage%1m-
ConnectionConnection Usage [Total]%1mDB connection usage rate (%)
DB AgeDB Age Maxage1mdatabase age (frozen XID) value
LockWait Lockscnt1mNumber of lock-waiting sessions (by DB)
LockWait Locks [Long Total]cnt1mNumber of sessions with long (300 seconds) lock waiting
LockWait Locks [Long]cnt1m-
LockWait Locks [Total]cnt1mNumber of sessions waiting due to lock occurrence
Long TransactionTransaction Time Max [Long]sec1m-
Long TransactionTransaction Time Max Total [Long]sec1mLong-running transaction time (minutes)
ReplicaApply Lag Timesec1mapply_lag time
ReplicaCheck No Replicationcnt1mcheck_no_replication value
ReplicaCheck Replicationstate1mcheck_replication_state value
SlowquerySlowqueriescnt1mNumber of SQL queries running for a long time (over 5 minutes)
StateInstance State [PID]PID1mpostgres process pid
TablespaceTablespace Usedbytes1mTablespace size
TablespaceTablespace Used [Total]bytes1m-
TablespaceTablespace Used Bytes [MB]bytes1mfilesystem directory usage (MB)
TablespaceTablespaces [Total]cnt1m-
Table. PostgreSQL (DBaaS) performance items
Reference
For the DB Instance performance items, refer to the Virtual Server performance items.

MariaDB(DBaaS)

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
ActivelockActive Lockscnt1mNumber of activelocks
ActivesssionActive Sessionscnt1mNumber of activesession
ActivesssionConnection Usage [Total]%1mDB connection session usage rate
ActivesssionConnectionscnt1mnumber of connections
ActivesssionConnections [MAX]cnt1mmax connected threads count
DatafileBinary Log Used [MB]bytes1mbinary log usage (MB)
DatafileData Directory Used [MB]bytes1mdatadir usage (MB)
DatafileOpen Filescnt1mNumber of DB files in open state
DatafileOpen Files [MAX]cnt1mNumber of DB files that can be opened
DatafileOpen Files Usage%1mDB file maximum count utilization
DatafileRelay Log Used [MB]bytes1mRelay log usage (MB)
StateInstance State [PID]PID1mmariadbd process pid
mysqld process pid(pre‑v10.5.2 version)
StateSafe PIDPID1mmariadbd_safe process pid
mysqld_safe process pid (prior to v10.5.2)
StateSlave Behind Master secondssec1mTime difference of Data between Master and Slave
(run only on slave)
TablespaceTablespace Usedbytes1mTablespace usage
TablespaceTablespace Used [Total]bytes1m-
TransactionRunning Threadscnt1mrunning thread count
TransactionSlowqueriescnt1mNumber of long-running SQL queries (over 5 minutes) (by DB)
TransactionSlowqueries [Total]cnt1mNumber of SQL queries running for a long time (over 5 minutes) (total)
TransactionTransaction Time [Long]sec1mTransaction maximum execution time (seconds)
TransactionWait Lockscnt1mNumber of sessions blocked for more than 60 seconds by lock
Table. MariaDB (DBaaS) performance items
Reference
For DB Instance performance metrics, refer to the performance metrics of Virtual Server.

MySQL(DBaaS)

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
ActivelockActive Lockscnt1mNumber of activelocks
ActivesssionActive Sessionscnt1mconnected threads count
ActivesssionConnection Usage [Total]%1mDB connection session usage rate
ActivesssionConnectionscnt1mnumber of connections
ActivesssionConnections [MAX]cnt1mmax connected threads count
DatafileBinary Log Used [MB]bytes1mbinary log usage (MB)
DatafileData Directory Used [MB]bytes1mdatadir usage (MB)
DatafileOpen Filescnt1mNumber of DB files in open state
DatafileOpen Files [MAX]cnt1mNumber of DB files that can be opened
DatafileOpen Files Usage%1mDB file maximum count utilization
DatafileRelay Log Used [MB]bytes1mRelay log usage (MB)
StateInstance State [PID]PID1mmysqld process pid
StateSafe PIDPID1msafe program PID
StateSlave Behind Master secondssec1mTime difference with master node (sec)
TablespaceTablespace Usedbytes1mTablespace usage
TablespaceTablespace Used [Total]bytes1mTablespace usage (total)
TransactionRunning Threadscnt1mrunning thread count
TransactionSlowqueriescnt1mNumber of long-running SQL queries (over 5 minutes) (by DB)
TransactionSlowqueries [Total]cnt1mNumber of SQL queries running for a long time (over 5 minutes) (total)
TransactionTransaction Time [Long]sec1mTransaction maximum execution time (seconds)
TransactionWait Lockscnt1mNumber of sessions blocked for more than 60 seconds by lock
Table. MySQL(DBaaS) performance items
Reference
Refer to the Virtual Server performance metrics for DB Instance performance items.

Microsoft SQL Server(DBaaS)

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
ActivelockActive Lockscnt1mNumber of activelocks
ActivesssionActive Sessionscnt1mNumber of activesession
ActivetransactionActive Transactions [Total]cnt1mNumber of active transactions
ConnectionConnected Userscnt1mNumber of users connected to the system
DatafileDatavolume Size [Free]bytes1mavailable space
DatafileDBFiles [Not Online]cnt1mRun a query to verify that all data files are in the ONLINE state.
DatafileTablespace Usedbytes1mData volume size
LockLock Processes [Blocked]cnt1mNumber of SQL processes blocked by other processes
LockLock Waits [Per Second]cnt1mLock wait count per second
SlowqueryBlocking Session IDID1mNumber of SQL queries running for a long time (over 5 minutes)
SlowquerySlowqueriescnt1mNumber of SQL queries running for a long time (over 5 minutes)
SlowquerySlowquery CPU Timems1mCPU time consumed by SQL execution that runs for a long time (over 5 minutes)
SlowquerySlowquery Execute Context IDID1mContext ID associated with the execution task of a SQL that runs for a long time (5 minutes or more)
SlowquerySlowquery Memory Usagebytes1mMemory usage consumed by the execution of SQL that runs for a long time (over 5 minutes)
SlowquerySlowquery Session IDID1mSession ID of SQL queries running for a long time (over 5 minutes)
SlowquerySlowquery Wait Duration Timems1mTotal wait time for wait type
StateInstance State [Cluster]state1mStatus during MSSQL cluster configuration
StateInstance State [PID]PID1msqlservr.exe process pid
StatePage IO Latch Wait Timems1mPage IO latch waits average wait time
TransactionTransaction Time [MAX]cnt1mLong-running (5 minutes or more) transaction
Table. Microsoft SQL Server (DBaaS) performance metrics

EPAS(DBaaS)

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
ActivelockAccess Exclusive Lockscnt1maccessexclusive lock count
ActivelockAccess Share Lockscnt1mNumber of accessshare locks
ActivelockActive Lockscnt1mNumber of activelocks
ActivelockActive Locks [Total]cnt1mactivelock count (total)
ActivelockExclusive Lockscnt1mexclusive lock count
ActivelockRow Exclusive Lockscnt1mrow exclusive lock count
ActivelockRow Share Lockscnt1mrow share lock count
ActivelockShare Lockscnt1mshare lock count
ActivelockShare Row Exclusive Lockscnt1mNumber of share row exclusive locks
ActivelockShare Update Exclusive Lockscnt1mNumber of share update exclusive locks
ActivesessionActive Sessionscnt1mNumber of active sessions
ActivesessionActive Sessions [Total]cnt1mTotal number of active sessions
ActivesessionIdel In Transaction Sessionscnt1mNumber of sessions in idle_in_transaction state
ActivesessionIdle In Transaction Sessions [Total]cnt1mTotal number of sessions in idle_in_transaction state
ActivesessionIdle Sessionscnt1mNumber of idle sessions
ActivesessionIdle Sessions [Total]cnt1mTotal number of idle sessions
ActivesessionWaiting Sessionscnt1mNumber of sessions in waiting state
ActivesessionWaiting Sessions [Total]cnt1mTotal number of sessions in waiting state
ConnectionConnection Usage%1mDB connection usage rate (%)
ConnectionConnection Usage [Total]%1mOverall DB connection usage (%)
ConnectionConnection Usage Per DB%1mDB connection usage rate (%) by DB
DB AgeDB Age Maxage1mdatabase age (frozen XID) value
LockWait Lockscnt1mNumber of sessions with long (300 seconds) lock waiting
LockWait Locks [Long Total]cnt1mTotal number of lock-waiting sessions (300 seconds)
LockWait Locks [Long]cnt1mNumber of sessions waiting due to lock occurrence
LockWait Locks [Total]cnt1mTotal number of sessions waiting due to lock occurrence
LockWait Locks Per DB [Total]cnt1mTotal number of sessions waiting due to lock occurrences per DB
Long TransactionTransaction Time Max [Long]sec1mLong-running transaction time (minutes)
Long TransactionTransaction Time Max Total [Long]sec1mLong-running transaction time (minutes)
ReplicaApply Lag Timesec1mapply_lag time
ReplicaCheck No Replicationcnt1mcheck_no_replication value
ReplicaCheck Replicationstate1mcheck_replication_state value
SlowquerySlowqueriescnt1mNumber of SQL queries running for a long time (over 5 minutes)
StateInstance state [PID]PID1medb-postgres process pid
TablespaceTablespace Used Bytes [MB]bytes1mfilesystem directory usage (MB)
TablespaceTablespaces [Total]cnt1mTotal Tablespace size
TablespaceTablespace Usedbytes1mSize of the tablespace in use
TablespaceTablespace Used [Total]bytes1mTotal size of the Tablespace in use
Table. EPAS (DBaaS) performance items

CacheStore(DBaaS)

Redis

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
StatsActive Defragmentation Keys [Hits]cnt1mNumber of keys after defragmentation
StatsActive Defragmentation Keys [Miss]cnt1mNumber of keys skipped in the active defragmentation removal process
StatsActive Defragmentationd [Hits]cnt1mNumber of value reassignments performed by the active defragmentation removal process
StatsActive Defragmentations [Miss]cnt1mNumber of value reallocations that were stopped, starting with the active defragmentation removal process
MemoryAllocated Bytes [OS]bytes1mNumber of bytes allocated by Redis and recognized by the operating system (resident set size)
MemoryAllocated Bytes [Redis]bytes1mTotal bytes allocated by Redis
PersistenceAOF Buffer Sizebytes1mAOF buffer size
PersistenceAOF File Size [Current]bytes1mAOF current file size
PersistenceAOF File Size [Lastest Startup]bytes1mAOF file size on recent start or rewrite
PersistenceAOF Rewrite Buffer Sizebytes1mAOF rewrite buffer size
PersistenceAOF Rewrite Current Timesec1mIf applicable, the time of the ongoing AOF rewrite operation
PersistenceAOF Rewrite Last Timesec1mFinal AOF rewrite operation time (seconds)
CommandstatsCallscnt1mNumber of calls that reached command execution (not rejected)
CommandstatsCalls [Failed]cnt1mNumber of failed calls
CommandstatsCalls [Rejected]cnt1mNumber of rejected calls
PersistenceChanges [Last Saved]cnt1mNumber of changes after the final dump
ClientsClient Output Buffer [MAX]cnt1mCurrent longest output list for client connections
ClientsClient Input Buffer [MAX]cnt1mMaximum input buffer for the current client connection
SentinelClients [Sentinel]cnt1mNumber of client connections (sentinel)
ReplicationConnected Slavescnt1mNumber of connected slaves
ClientsConnections [Blocked]cnt1mNumber of clients pending blocking calls (BLPOP, BRPOP, BRPOPLPUSH)
ClientsConnections [Current]cnt1mNumber of client connections (excluding slave connections)
PersistenceCopy On Write Allocated Size [AOF]bytes1mCOW allocation size during final RBD save operation
PersistenceCopy On Write Allocated Size [RDB]bytes1mCOW allocation size during final RBD save operation
CommandstatsCPU Time [Average]cnt1mAverage CPU used per command execution
CommandstatsCPU Time [Total]usec1mTotal CPU time used by these commands
CPUCPU Usage [System Process]%1mSystem CPU used by background processes
CPUCPU Usage [System]%1mSystem CPU used by the Redis server
CPUCPU Usage [User Process]%1mUser CPU used by background processes
CPUCPU Usage [User]%1mSystem CPU used by background processes
MemoryDataset Usedbytes1mDataset size
DiskDisk Usedbytes1mdatadir usage
StatsEvicted Keyscnt1mNumber of evicted keys caused by the maxmemory limit
PersistenceFsyncs [Delayed]cnt1mDelayed fsync counter
PersistenceFsyncs [Pending]cnt1mNumber of pending fsync operations in the background I/O queue (format: bytes)
StatsFull Resyncscnt1mNumber of full resynchronizations with the slave
StatsKeys [Expired]cnt1mTotal number of key expiration events
KeyspaceKeys [Keyspace]cnt1mNumber of keys in the key space
StatsLastest Fork Duration Timeusec1mRecent fork operation time (microseconds)
StatsLookup Keys [Hit]cnt1mNumber of successful key lookups in the main dictionary
StatsLookup Keys [Miss]cnt1mNumber of failed key lookups in the main dictionary
MemoryLua Engine Memory Usedbytes1mMemory used by the Lua engine
ReplicationMaster Last Interaction Time Agosec1mElapsed time (seconds) since the final interaction with the master
ReplicationMaster Last Interaction Time Ago [Sync]sec1mElapsed time (seconds) since the final interaction with the master
ReplicationMaster Offsetpid1mCurrent replication offset of the server
ReplicationMaster Second Offsetpid1mOffset until the replica ID is accepted
ReplicationMaster Sync Left Bytesbytes1mRemaining bytes before synchronization completes
MemoryMemory Fragmentation Rate%1mused_memory_rss and used_memory ratio
MemoryMemory Fragmentation Rate [Allocator]%1mfragmentation ratio
MemoryMemory Fragmentation Usedbytes1mBytes between used_memory_rss and used_memory
MemoryMemory Fragmentation Used [Allocator]bytes1mresident byte
MemoryMemory Max Valuebytes1mMemory limit
MemoryMemory Resident [Allocator]bytes1mresident memory
MemoryMemory RSS Rate [Allocator]%1mresident ratio
MemoryMemory Used [Active]bytes1mActive memory
MemoryMemory Used [Allocated]bytes1mAllocated memory
MemoryMemory Used [Resident]bytes1mresident byte
StatsNetwork In Bytes [Total]bytes1mTotal network input
StatsNetwork Out Bytes [Total]bytes1mTotal network output
StatsNetwork Read Ratecnt1mNetwork read speed (KB/sec)
StatsNetwork Write Ratecnt1mNetwork write speed (KB/sec)
StatsPartial Resync Requests [Accepted]cnt1mNumber of accepted partial resynchronization requests
StatsPartial Resync Requests [Denied]cnt1mNumber of re-sync requests for rejected parts
MemoryPeak Memory Consumedbytes1mMaximum memory used by Redis
StatsProcessed Commandscnt1mNumber of commands processed per second
StatsProcessed Commands [Total]cnt1mTotal number of processed commands
StatsPub/Sub Channelscnt1mGlobal count of pub/sub channels with client subscriptions
StatsPub/Sub Patternscnt1mGlobal count of publish/subscribe pattern with client subscriptions
PersistenceRDB Saved Duration Time [Current]sec1mIf applicable, the time of the ongoing RDB save operation
PersistenceRDB Saved Duration Time [Last]sec1mFinal RDB save operation time (seconds)
StatsReceived Connections [Total]cnt1mTotal number of received connections
StatsRejected Connections [Total]cnt1mTotal number of rejected connections
ReplicationReplication Backlog Actove Countcnt1mReplication backlog enable flag
ReplicationReplication Backlog Master Offsetcnt1mMaster offset of the replication backlog buffer
ReplicationReplication Backlog Sizebytes1mData size of the replication backlog buffer (bytes)
ReplicationReplication Backlog Size [Total]bytes1mTotal size of the replication backlog buffer (bytes)
ReplicationSlave Prioritycnt1mPriority of instances as a fault handling target
ReplicationSlave Replication Offsetpid1mReplication offset of the slave instance
SlowlogSlow Operationscnt1mNumber of slow tasks
StatsSockets [MIGRATE]cnt1mNumber of sockets opened for migration
StatsTracked Keys [Expiry]cnt1mNumber of keys tracked for expiration (applicable only to writable slaves)
StateInstance Status [PID]PID1mredis-server process pid
StateSentinel Status [PID]PID1msentinel process pid
Table. Redis performance metrics
Reference
For DB Instance performance metrics, see the Virtual Server performance metrics.

Valkey

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
StatsActive Defragmentation Keys [Hits]cnt1mNumber of keys after defragmentation
StatsActive Defragmentation Keys [Miss]cnt1mNumber of keys skipped in the active defragmentation removal process
StatsActive Defragmentationd [Hits]cnt1mNumber of value reassignments performed by the active defragmentation removal process
StatsActive Defragmentations [Miss]cnt1mNumber of value reallocations that were stopped, starting with the active defragmentation removal process
MemoryAllocated Bytes [OS]bytes1mNumber of bytes allocated by Valkey and recognized by the operating system (resident set size)
MemoryAllocated Bytes [Valkey]bytes1mTotal bytes allocated by Valkey
PersistenceAOF Buffer Sizebytes1mAOF buffer size
PersistenceAOF File Size [Current]bytes1mAOF current file size
PersistenceAOF File Size [Lastest Startup]bytes1mAOF file size on recent start or rewrite
PersistenceAOF Rewrite Buffer Sizebytes1mAOF rewrite buffer size
PersistenceAOF Rewrite Current Timesec1mIf applicable, the time of the ongoing AOF rewrite operation
PersistenceAOF Rewrite Last Timesec1mFinal AOF rewrite operation time (seconds)
CommandstatsCallscnt1mNumber of calls that reached command execution (not rejected)
CommandstatsCalls [Failed]cnt1mNumber of failed calls (Valkey 6.2-rc2)
CommandstatsCalls [Rejected]cnt1mRejected call count (Valkey 6.2-rc2)
PersistenceChanges [Last Saved]cnt1mNumber of changes after the final dump
ClientsCleint Output Buffer [MAX]cnt1mCurrent longest output list for client connections
ClientsClient Input Buffer [MAX]cnt1mMaximum input buffer for current client connections (Valkey 5.0)
SentinelClients [Sentinel]cnt1mNumber of client connections (sentinel)
ReplicationConnected Slavescnt1mNumber of connected slaves
ClientsConnections [Blocked]cnt1mNumber of clients pending blocking calls (BLPOP, BRPOP, BRPOPLPUSH)
ClientsConnections [Current]cnt1mNumber of client connections (excluding slave connections)
PersistenceCopy On Write Allocated Size [AOF]bytes1mCOW allocation size during final RBD save operation
PersistenceCopy On Write Allocated Size [RDB]bytes1mCOW allocation size during final RBD save operation
CommandstatsCPU Time [Average]cnt1mAverage CPU used per command execution
CommandstatsCPU Time [Total]usec1mTotal CPU time used by these commands
CPUCPU Usage [System Process]%1mSystem CPU used by background processes
CPUCPU Usage [System]%1mSystem CPU used by the Valkey server
CPUCPU Usage [User Process]%1mUser CPU used by background processes
CPUCPU Usage [User]%1mSystem CPU used by background processes
MemoryDataset Usedbytes1mDataset size
DiskDisk UsedMB1mdatadir usage
StatsEvicted Keyscnt1mNumber of evicted keys caused by the maxmemory limit
PersistenceFsyncs [Delayed]cnt1mDelayed fsync counter
PersistenceFsyncs [Pending]cnt1mNumber of pending fsync operations in the background I/O queue (format: bytes)
StatsFull Resyncscnt1mNumber of full resynchronizations with the slave
StatsKeys [Expired]cnt1mTotal number of key expiration events
KeyspaceKeys [Keyspace]cnt1mNumber of keys in the key space
StatsLastest Fork Duration Timeusec1mRecent fork operation time (microseconds)
StatsLookup Keys [Hit]cnt1mNumber of successful key lookups in the main dictionary
StatsLookup Keys [Miss]cnt1mNumber of failed key lookups in the main dictionary
MemoryLua Engine Memory Usedbytes1mMemory used by the Lua engine
ReplicationMaster Last Interaction Time Agosec1mElapsed time (seconds) since the final interaction with the master
ReplicationMaster Last Interaction Time Ago [Sync]sec1mElapsed time (seconds) since the final interaction with the master
ReplicationMaster Offsetpid1mCurrent replication offset of the server
ReplicationMaster Second Offsetpid1mOffset until the replica ID is accepted
ReplicationMaster Sync Left Bytesbytes1mRemaining bytes before synchronization completes
MemoryMemory Fragmentation Rate%1mused_memory_rss and used_memory ratio
MemoryMemory Fragmentation Rate [Allocator]%1mfragmentation ratio
MemoryMemory Fragmentation Usedbytes1mBytes between used_memory_rss and used_memory
MemoryMemory Fragmentation Used [Allocator]bytes1mresident byte
MemoryMemory Max Valuebytes1mMemory limit
MemoryMemory Resident [Allocator]bytes1mresident memory
MemoryMemory RSS Rate [Allocator]%1mresident ratio
MemoryMemory Used [Active]bytes1mActive memory
MemoryMemory Used [Allocated]bytes1mAllocated memory
MemoryMemory Used [Resident]bytes1mresident byte
StatsNetwork In Bytes [Total]bytes1mTotal network input
StatsNetwork Out Bytes [Total]bytes1mTotal network output
StatsNetwork Read Ratekbps1mNetwork read speed (KB/sec)
StatsNetwork Write Ratekbps1mNetwork write speed (KB/sec)
StatsPartial Resync Requests [Accepted]cnt1mNumber of accepted partial resynchronization requests
StatsPartial Resync Requests [Denied]cnt1mNumber of re-sync requests for rejected parts
MemoryPeak Memory Consumedbytes1mMaximum memory used by Valkey
StatsProcessed Commandscnt1mNumber of commands processed per second
StatsProcessed Commands [Total]cnt1mTotal number of processed commands
StatsPub/Sub Channelscnt1mGlobal count of pub/sub channels with client subscriptions
StatsPub/Sub Patternscnt1mGlobal count of publish/subscribe pattern with client subscriptions
PersistenceRDB Saved Duration Time [Current]sec1mIf applicable, the time of the ongoing RDB save operation
PersistenceRDB Saved Duration Time [Last]sec1mFinal RDB save operation time (seconds)
StatsReceived Connections [Total]cnt1mTotal number of received connections
StatsRejected Connections [Total]cnt1mTotal number of rejected connections
ReplicationReplication Backlog Active Countcnt1mReplication backlog enable flag
ReplicationReplication Backlog Master Offsetcnt1mMaster offset of the replication backlog buffer
ReplicationReplication Backlog Sizebytes1mData size of the replication backlog buffer
ReplicationReplication Backlog Size [Total]bytes1mTotal size of the replication backlog buffer
ReplicationSlave Prioritycnt1mPriority of instances as a fault handling target
ReplicationSlave Replication Offsetpid1mReplication offset of the slave instance
SlowlogSlow Operationscnt1mNumber of slow tasks
StatsSockets [MIGRATE]cnt1mNumber of sockets opened for migration
StatsTracked Keys [Expiry]cnt1mNumber of keys tracked for expiration (applicable only to writable slaves)
StateInstance State [PID]PID1mValkey-server process PID
StateSentinel State [PID]PID1mSentinel process PID
Table. Valkey performance metrics
reference
Refer to the Virtual Server performance items for DB Instance performance metrics.

Data Analytics type

Event Streams

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
BrokerConnections [Zookeeper Client]cnt1mNumber of ZooKeeper connections
BrokerFailed [Client Fetch Request]cnt1mClient fetch request processing failure count
BrokerFailed [Produce Request]cnt1mProcucer request processing failure count
BrokerIncomming Messagescnt1mNumber of messages received by the broker
BrokerLeader Electionscnt1mLeader Election occurrence count
BrokerLeader Elections [Unclean]cnt1mNumber of Unclean Leader Election occurrences
BrokerLog Flushescnt1mNumber of log flush occurrences
BrokerNetwork In Bytesbytes1mTotal bytes received by the Topic
BrokerNetwork Out Bytesbytes1mTotal bytes transmitted by the Topic
BrokerRejected Bytesbytes1mTotal bytes rejected by the Topic
BrokerRequest Queue Lengthcnt1mRequest queue size
BrokerZookeeper Sessions [Closed]cnt1mZooKeeper closed sessions per second
BrokerZookeeper Sessions [Expired]cnt1mZooKeeper expired sessions per second
BrokerZookeeper Sessions [Readonly]cnt1mZooKeeper read‑only sessions per second
BrokerIncomming Messages Rate [Topic]cnt1mNumber of received messages per topic
BrokerIncomming Byte Rate [Second]bytes1mper second Incomming data
BrokerOutgoing Byte Rate [Second]bytes1mOutgoing data per second
BrokerRejected Byte Rate [Second]bytes1mBytes rejected per second
DiskDisk Usedbytes1mDatadir usage
StateAKHQ State [PID]PID1makhq process pid
StateInstance State [PID]PID1mkafka process pid
StateZookeeper State [PID]PID1mzookeeper process pid
Table. Event Streams performance metrics

Search Engine

Elasticsearch

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
ClusterShardscnt1mNumber of cluster shards
ClusterShards [Primary]cnt1mNumber of primary shards in the cluster
ClusterIndex [Total]cnt1mNumber of clustered indexes
ClusterLicense Expiry Date [ms]ms1mLicense expiration date [milisecond]
ClusterLicense Statusstate1mLicense status
ClusterLicense Typetype1mLicense type
FileSystemDisk Usagebytes1mdatadir usage
NodeDocuments [Deleted]cnt1mTotal number of deleted documents
NodeDocuments [Existing]cnt1mTotal number of existing documents
NodeFilesystem Bytes [Available]bytes1mAvailable file systems
NodeFilesystem Bytes [Free]bytes1mAvailable file system
NodeFilesystem Bytes [Total]bytes1mTotal file system
NodeJVM Heap Used [Init]bytes1mHeap init used by JVM (bytes)
NodeJVM Heap Used [MAX]bytes1mHeap max used by JVM (bytes)
NodeJVM Non Heap Used [Init]bytes1minit(bytes) excluding the heap used by the JVM
NodeJVM Non Heap Used [MAX]bytes1mmax (bytes) excluding the heap used by the JVM
NodeSegmentscnt1mTotal number of segments
NodeSegments Bytesbytes1mTotal size of the segment
NodeStore Bytesbytes1mTotal size of the repository
StateInstance state [PID]PID1mElasticsearch process pid
TaskQueue Timems1mQueue time
KibanaKibana state [PID]PID1mKibana process pid
KibanaKibana Connectionscnt1mconnection
KibanaKibana Memory Heap Allocated [Limit]bytes1mMaximum old space size allocated to the Node.js process
KibanaKibana Memory Heap Allocated [Total]bytes1mMemory
KibanaKibana Memory Heap Usedbytes1mMemory
KibanaKibana Process Uptimems1mProcess
KibanaKibana Requests [Disconnected]cnt1mRequest count metric
KibanaKibana Requests [Total]cnt1mRequest count metric
KibanaKibana Response Time [Avg]ms1mResponse time metric
KibanaKibana Response Time [MAX]ms1mResponse time metric
Table. Elasticsearch performance metrics

Opensearch

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
StateCluster statestate1mCluster status
ClusterNodescnt1mNumber of nodes in the cluster
ClusterData nodescnt1mNumber of data nodes in the cluster
ClusterPending taskscnt1mNumber of pending tasks
ShardShards [active]cnt1mActive piece count
ShardShards [active_primary]cnt1mNumber of active primary fragments
ShardShards [initializing]cnt1mInitial shard count
ShardShards [relocating]cnt1mPrevious piece count
ShardShards [unassigned]cnt1mNumber of unallocated fragments
ThreadThread Queue Count [search]cnt1mNumber of search tasks in the queue
ThreadThread Queue Count [refresh]cnt1mNumber of refresh tasks in the queue
ThreadThread Queue Count [write]cnt1mNumber of write operations in the queue
ThreadThread Queue Count [get]cnt1mNumber of jobs fetched from the queue
ThreadThread Queue Count [snapshot]cnt1mNumber of snapshot jobs in the queue
ThreadThread Queue Count [flush]cnt1mNumber of flush operations in the queue
ThreadThread Queue Count [force_merge]cnt1mNumber of force_merge tasks in the queue
SystemCPU usage%1mCPU usage
SystemMemory usagebytes1mUsed memory
SystemDisk availablebytes1mDisk Available
DocumentsDocuments indexing ratecnt1mNumber of indexed documents
DocumentsDocuments indexing rate [Delta]cnt1mNumber of indexed documents (delta value)
DocumentsIndexing latencysec1mTime taken to index documents
DocumentsIndexing latency [Delta]sec1mTime taken to index the document (delta value)
DocumentsSearch ratecnt1mNumber of search queries
DocumentsSearch rate [Delta]cnt1mNumber of search queries (delta value)
DocumentsSearch latencysec1mTime taken during the query
DocumentsSearch latency [Delta]sec1mTime taken during the query (delta value)
DocumentsDocument count (with replicas)cnt1mTotal number of documents
DocumentsDocument deleting ratecnt1mNumber of deleted documents
DocumentsDocument deleting rate [Delta]cnt1mNumber of deleted documents (delta value)
DocumentsDocument merging ratecnt1mNumber of merged documents
DocumentsDocument merging rate [Delta]cnt1mNumber of merged documents (delta value)
JVMHeap usedbytes1mMemory used in the heap
JVMGC count [young]cnt1mNumber of young GC collections
JVMGC count [young] [Delta]cnt1mYoung GC collection count (delta value)
JVMGC count [G1]cnt1mG1 GC collection count
JVMGC count [G1] [Delta]cnt1mG1 GC collection count (delta value)
JVMGC count [old]cnt1mNumber of previous GC collections
JVMGC count [old] [Delta]cnt1mPrevious GC collection count (delta value)
JVMGC time [young]cnt1mTime spent on young GC collection
JVMGC time [young] [Delta]cnt1mTime spent for young GC collection (delta value)
JVMGC time [G1]cnt1mTime spent on G1 GC collection
JVMGC time [G1] [Delta]cnt1mTime spent on G1 GC collection (delta value)
JVMGC time [old]cnt1mTime spent on old GC collections
JVMGC time [old] [Delta]cnt1mTime spent on old GC collections (delta value)
StateInstance state [PID]PID1mOpensearch process PID
StateDashboard state [PID]PID1mDashboard process PID
Table. Opensearch performance items

Vertica(DBaaS)

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
StateInstance State [PID]state1mVertica process PID
ActivelockActive Lockscnt1mActive Locks count
ActivesessionActive Sessionscnt1mNumber of Active Sessions
TablespaceData Tablespace UsedMB1mData, Temp Tablespace usage
TablespaceCatalog Tablespace UsedMB1mCatalog Tablespace Usage
Table. Vertica (DBaaS) performance metrics
Reference
Refer to the Virtual Server performance items for DB Instance performance metrics.

Container type

Kubernetes Engine

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
ClusterCluster Namespaces [Active]cnt5mNumber of namespaces in active state
ClusterCluster Namespaces [Total]cnt5mTotal number of namespaces in the cluster
ClusterCluster Nodes [Ready]cnt5mNumber of nodes in READY state
ClusterCluster Nodes [Total]cnt5mTotal number of nodes in the cluster
ClusterCluster Pods [Failed]cnt5mNumber of failed-state pods in the cluster
ClusterCluster Pods [Pending]cnt5mNumber of pending pods in the cluster
ClusterCluster Pods [Running]cnt5mNumber of pods in running state within the cluster
ClusterCluster Pods [Succeeded]cnt5mNumber of succeeded pods in the cluster
ClusterCluster Pods [Unknown]cnt5mNumber of pods in unknown state within the cluster
ClusterInstance Statestate5mcluster status
NamespaceNamespace Pods [Failed]cnt5mNumber of failed-state pods in a namespace
NamespaceNamespace Pods [Pending]cnt5mNumber of pending pods in the namespace
NamespaceNamespace Pods [Running]cnt5mNumber of running pods in a namespace
NamespaceNamespace Pods [Succeeded]cnt5mNumber of succeeded pods in the namespace
NamespaceNamespace Pods [Unknown]cnt5mNumber of unknown-state pods in the namespace
NamespaceNamespace GPU Clock FrequencyMHz5mSM clock frequency in the Namespace
NamespaceNamespace GPU Memory Usage%5mMemory utilization in Namespace
NodeNode CPU Size [Allocatable]cnt5mNode allocatable CPU
NodeNode CPU Size [Capacity]cnt5mCPU capacity within the node
NodeNode CPU Usage%5mCPU usage on the node
NodeNode CPU Usage [Request]%5mCPU request_ratio within node
NodeNode CPU Usedstate5mCPU utilization within the node
NodeNode Filesystem Usage%5mFS usage within node
NodeNode Memory Size [Allocatable]bytes5mmemory allocatable within the node
NodeNode Memory Size [Capacity]bytes5mNode memory utilization
NodeNode Memory Usage%5mNode memory utilization
NodeNode Memory Usage [Request]%5mmemory request_ratio within the node
NodeNode Memory Workingsetbytes5mmemory working set within the node
NodeNode Network In Bytesbytes5mNode network rx bytes
NodeNode Network Out Bytesbytes5mNode network tx bytes
NodeNode Network Total Bytesbytes5mNode network total bytes
NodeNode Pods [Failed]cnt5mNumber of pods in failed state within a node
NodeNode Pods [Pending]cnt5mNumber of pending pods in the node
NodeNode Pods [Running]cnt5mNumber of running pods per node
NodeNode Pods [Succeeded]cnt5mNumber of succeeded pods in the node
NodeNode Pods [Unknown]cnt5mNumber of pods in unknown state on the node
PodPod CPU Usage [Limit]%5mCPU usage_limit_ratio within the pod
PodPod CPU Usage [Request]%5mCPU request_ratio within the pod
PodPod CPU Usagemc5mCPU usage within the pod
PodPod Memory Usage [Limit]%5mmemory usage_limit_ratio in the pod
PodPod Memory Usage [Request]%5mmemory request_ratio in pod
PodPod Memory Usagebytes5mMemory usage within the pod
PodPod Network In Bytesbytes5mnetwork rx bytes in pod
PodPod Network Out Bytesbytes5mnetwork tx bytes in pod
PodPod Network Total Bytesbytes5mNetwork total bytes in pod
PodPod Restart Containerscnt5mcontainer restart count in pod
WorkloadWorkload Pods [Running]cnt5m-
Table. Kubernetes Engine performance items

Container Registry

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
Container RegistryImage Pulls [Denied]cnt1mNumber of rejected Image Tag (digest) Pulls
Container RegistryImage Pushs [Allowed]cnt1mAllowed Image Tag (digest) Push count
Container RegistryImage Pushs [Denied]cnt1mNumber of rejected Image Tag (digest) Pushes
Container RegistryImage Scans[Allowed]cnt1mAllowed Image Tag (digest) Scan count
Container RegistryImage Scans [Denied]cnt1mNumber of rejected Image Tag (digest) scans
Container RegistryImage Tags [Deleted]cnt1mNumber of deleted Image Tag (digest)
Container RegistryImages [Created]cnt1mNumber of generated images
Container RegistryImages [Deleted]cnt1mNumber of deleted images
Container RegistryLogins [Allowed]cnt1mNumber of allowed Registry Logins
Container RegistryLogins [Denied]cnt1mNumber of denied Registry Logins
Container RegistryRepositories [Created]cnt1mNumber of created repositories
Container RegistryRepositories [Deleted]cnt1mNumber of deleted repositories
StateInstance Statestate1mCheck status
Table. Container Registry performance metrics

Networking type

Internet Gateway

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
Internet GatewayNetwork In Total Bytes [Internet Delta]bytes5mInternet Gateway → Cumulative traffic volume toward VPC for 5 minutes (Internet)
※ Traffic bps average conversion formula: cumulative traffic volume (bytes) / 300 (seconds) * 8 (bits)
Internet GatewayNetwork In Total Bytes [Internet]bytes5mrx bytes total
Internet GatewayNetwork Out Total Bytes [Internet Delta]bytes5mVPC → cumulative traffic volume toward the Internet Gateway over 5 minutes (Internet)
※ Traffic bps average conversion formula: cumulative traffic volume (bytes) / 300 (seconds) * 8 (bits)
Internet GatewayNetwork Out Total Bytes [Internet]bytes5mtx bytes total
Table. Internet Gateway performance metrics

Load Balancer(OLD)

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
Load BalancerCurrent Connectioncnt5mCurrent number of connections
Load BalancerTotal Connectioncnt5mTotal number of connections
Load BalancerTotal Connection [Delta]cnt5mTotal number of connections (delta value)
Load BalancerNetwork In Bytesbytes5min bytes
Load BalancerNetwork In Bytes [Delta]bytes5mClient → Load Balancer cumulative traffic volume over 5 minutes
※ Traffic bps average conversion formula: cumulative traffic volume (bytes) / 300 (seconds) * 8 (bit)
Load BalancerNetwork Out Bytesbytes5mout bytes
Load BalancerNetwork Out Bytes [Delta]bytes5mCumulative traffic volume over 5 minutes from Load Balancer to Client
※ Traffic bps average conversion formula: cumulative traffic volume (bytes) / 300 (seconds) * 8 (bits)
Load BalancerInstance Statestate5mLoad Balancer status
Table. Load Balancer performance items

Load Balancer Listener(OLD)

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
ListenerConnections [Current]cnt5mCurrent number of connections
ListenerConnections [Total Delta]cnt5mtotal connection count (delta value)
ListenerConnections [Total]cnt5mtotal connection count
ListenerInstance Statestate5mLB Listener status
ListenerNetwork In Bytesbytes5min bytes
ListenerNetwork In Bytes [Delta]bytes5mCumulative traffic volume over 5 minutes from Client to Load Balancer
※ Traffic bps average conversion formula: cumulative traffic volume (bytes) / 300 (seconds) * 8 (bits)
ListenerNetwork Out Bytesbytes5mout bytes
ListenerNetwork Out Bytes [Delta]bytes5mLoad Balancer → Client cumulative traffic volume over 5 minutes
※ Traffic bps average conversion formula: cumulative traffic volume (bytes) / 300 (seconds) * 8 (bits)
Table. Load Balancer Listener performance metrics

Direct Connect

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
Direct ConnectNetwork In Bytesbytes5mCumulative traffic volume from Direct Connect → VPC
Direct ConnectNetwork In Bytes [Delta]bytes5mDirect Connect → VPC cumulative traffic volume over 5 minutes
※ Traffic bps average conversion formula: Cumulative traffic volume (bytes) / 300 (seconds) * 8 (bits)
Direct ConnectNetwork Out Bytesbytes5mCumulative traffic volume from VPC to Direct Connect
Direct ConnectNetwork Out Bytes [Delta]bytes5mVPC → Direct Connect cumulative traffic volume over 5 minutes
※ Traffic bps average conversion formula: cumulative traffic volume (bytes) / 300 (seconds) * 8 (bits)
Table. Direct Connect performance items

Load Balancer

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
StateInstance Statestate5mLB status
Load BalancerCurrent Connectioncnt5mCurrent number of connections
Load BalancerTotal L4 Connectioncnt5mTotal L4 Connection count
Load BalancerTotal L7 Connectioncnt5mTotal number of L7 connections
Load BalancerTotal TCP Connectioncnt5mTotal number of TCP connections
Load BalancerTotal Connectioncnt5mTotal number of connections
Load BalancerBytes processed in forward directionbytes5mFull‑duplex Network Byte
Load BalancerPackets processed in forward directioncnt5mBidirectional Network packet
Load BalancerBytes processed in reverse directionbytes5mReverse Network Byte
Load BalancerPackets processed in reverse directioncnt5mReverse Network packet
Load BalancerTotal failure actionscnt5mTotal number of failures
Load BalancerCurrent Requestcnt5mCurrent request count
Load BalancerCurrent responsecnt5mCurrent Response count
Load BalancerTotal Requestcnt5mTotal number of requests
Load BalancerTotal Request Successcnt5mTotal number of successful requests
Load BalancerPeak Connectioncnt5mMaximum number of connections
Load BalancerCurrent Connection Rate%5mCurrent SSL Connection rate
Load BalancerLast response timems5mLast response time
Load BalancerFastest response timems5mShortest response time
Load BalancerSlowest response timems5mMaximum response time
Load BalancerCurrent SSL Connectioncnt5mCurrent number of SSL connections
Load BalancerTotal SSL Connectioncnt5mTotal number of SSL connections
Load BalancerBytes processed in forward direction [Delta]bytes5mForward Network  Byte (delta value)
Load BalancerPackets processed in forward direction [Delta]cnt5mForward Network packet (delta value)
Load BalancerBytes processed in reverse direction [Delta]bytes5mReverse Network Byte (delta value)
Load BalancerPackets processed in reverse direction [Delta]cnt5mReverse Network packet (delta value)
Table. Load Balancer performance items

Load Balancer Listener

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
StateInstance Statestate5mLB status
Load BalancerCurrent Connectioncnt5mCurrent number of connections
Load BalancerTotal L4 Connectioncnt5mTotal L4 Connection count
Load BalancerTotal L7 Connectioncnt5mTotal number of L7 connections
Load BalancerTotal TCP Connectioncnt5mTotal number of TCP connections
Load BalancerTotal Connectioncnt5mTotal number of connections
Load BalancerBytes processed in forward directionbytes5mFull‑duplex Network Byte
Load BalancerPackets processed in forward directioncnt5mBidirectional Network packet
Load BalancerBytes processed in reverse directionbytes5mReverse Network Byte
Load BalancerPackets processed in reverse directioncnt5mReverse Network packet
Load BalancerTotal failure actionscnt5mTotal number of failures
Load BalancerCurrent Requestcnt5mCurrent request count
Load BalancerCurrent responsecnt5mCurrent Response count
Load BalancerTotal Requestcnt5mTotal number of requests
Load BalancerTotal Request Successcnt5mTotal number of successful requests
Load BalancerPeak Connectioncnt5mMaximum number of connections
Load BalancerCurrent Connection Rate%5mCurrent SSL Connection rate
Load BalancerLast response timems5mLast response time
Load BalancerFastest response timems5mShortest response time
Load BalancerSlowest response timems5mMaximum response time
Load BalancerCurrent SSL Connectioncnt5mCurrent number of SSL connections
Load BalancerTotal SSL Connectioncnt5mTotal number of SSL connections
Load BalancerBytes processed in forward direction [Delta]bytes5mForward Network  Byte (delta value)
Load BalancerPackets processed in forward direction [Delta]cnt5mForward Network packet (delta value)
Load BalancerBytes processed in reverse direction [Delta]bytes5mReverse Network Byte (delta value)
Load BalancerPackets processed in reverse direction [Delta]cnt5mReverse Network packet (delta value)
Table. Load Balancer Listener performance metrics

Load Balancer Server Group

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
Server GroupInstance Statestate5mLB Server Group status
Server GroupPeak Connectioncnt5mMaximum connections per server group
Server GroupHealthy hostcnt5mNumber of healthy hosts in server group
Server GroupUnhealthy hostcnt5mNumber of abnormal hosts in server group
Server GroupRequest Countcnt5mNumber of requests
Server GroupResponse Countcnt5mResponse count
Server Group2xx Response Countcnt5m2xx response count
Server Group3xx Response Countcnt5mNumber of 3xx responses
Server Group4xx Response Countcnt5m4xx response count
Server Group5xx Response Countcnt5mNumber of 5xx responses
Table. Load Balancer Server Group performance metrics

Cloud WAN

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
StateInstance Statestate10mAttachment connection status
AttachmentNetwork in bytesbytes10min bytes(Inbound traffic usage per interval)
AttachmentNetwork out bytesbytes10mOut bytes(Outbound traffic usage per interval)
AttachmentNetwork In Packets [Dropped]cnt10min Dropped Packet count (number of dropped packets per interval)
AttachmentNetwork Out Packets [Dropped]cnt10mOut Dropped Packet count (number of dropped packets per interval)
AttachmentNetwork In Packets [Unicast]cnt10min Unicast Packet count (number of Unicast packets per cycle)
AttachmentNetwork Out Packets [Unicast]cnt10mOut Unicast Packet count (Unicast packets per cycle)
AttachmentNetwork In Packets [Broadcast]cnt10min Broadcast Packet count (number of Broadcast packets per cycle)
AttachmentNetwork Out Packets [Broadcast]cnt10mOut Broadcast Packet count (number of broadcast packets per cycle)
AttachmentNetwork In Packets [Multicast]cnt10min Multicast Packet count (Multicast packets per cycle)
AttachmentNetwork Out Packets [Multicast]cnt10mOut Multicast Packet count (Multicast packet count per cycle)
AttachmentNetwork In Error Packetscnt10min Error Packet count (number of received error packets per cycle)
AttachmentNetwork Out Error Packetscnt10mOut Error Packet count (number of transmitted error packets per cycle)
Table. Cloud WAN performance metrics

Global CDN

Performance Item Group NamePerformance item namecollection unitCollection intervalExplanation
Global CDNInstance Statestate5mGlobal CDN status
Global CDNData Transfer Bytesbytes5mData transfer volume transmitted via CDN service (originBytes)
Global CDNRequests [Total]cnt5mNumber of service requests (cases) received by the CDN service (originHits)
Table. Global CDN performance metrics

2.9 - Appendix C. Service-specific Status Checks

Compute type

Virtual Server

Performance item nameExplanationvalue
Instance State [Basic]Instance statusNOSTATE, RUNNING, BLOCKED, PASUED, SHUTDOWN, SHUTOFF, CRASHED, PMSUSPENDED, LAST
Table. Virtual Server status check

GPU Server

Performance item namedescriptionvalue
Instance State [Basic]Instance statusNOSTATE RUNNING, BLOCKED, PASUED, SHUTDOWN, SHUTOFF, CRASHED, PMSUSPENDED, LAST
Table. GPU Server Status Check

Bare Metal Server

Performance item namedescriptionvalue
N/AN/AN/A
Table. Bare Metal Server status check
Caution
Bare Metal Server does not provide status information through Cloud Monitoring.

Multi-node GPU Cluster [Cluster Fabric]

Performance item namedescriptionvalue
N/AN/AN/A
Table. Multi-node GPU Cluster [Cluster Fabric] Status Check
Caution
Multi-node GPU Cluster [Cluster Fabric] does not provide status information through Cloud Monitoring.

Multi-node GPU Cluster [Node]

Performance item namedescriptionvalue
N/AN/AN/A
Table. Multi-node GPU Cluster [Node] Status Check
Caution
Multi-node GPU Cluster [Node] does not provide status information through Cloud Monitoring.

Storage type

File Storage

Performance item namedescriptionvalue
Instance StateFile Storage Volume Status1: When Online
0: Other status values (Offline)
Table. File Storage status check

Object Storage

Performance item namedescriptionvalue
N/AN/AN/A
Table. Object Storage status check
Caution
Object Storage does not provide status information through Cloud Monitoring.

Block Storage(BM)

Performance item nameExplanationvalue
Instance StateBlockstorage volume status1: running (normal)
* 0: down (abnormal)
Table. Block Storage (BM) status check

Block Storage(VM)

Performance item namedescriptionvalue
Instance StateBlockstorage volume status1: running (normal)
* 0: down (abnormal)
Table. Block Storage(VM) status check

Database type

PostgreSQL(DBaaS)

Performance item nameExplanationvalue
Instance State [PID]postgres process PIDPID: postgres when the process exists
* -1: when the process does not exist
Table. PostgreSQL (DBaaS) status check

MariaDB(DBaaS)

Performance item namedescriptionvalue
Safe PIDmariadb_safe process PIDPID: mariadb_safe when the process exists
-1: when the process does not exist
Instance State [PID]mariadb process PIDPID: mariadb if the process exists
* -1: if the process does not exist
Table. MariaDB (DBaaS) status check

MySQL(DBaaS)

Performance item namedescriptionvalue
Instance State [PID]mysqld process PIDPID: mysqld process exists
-1: process does not exist
Table. MySQL(DBaaS) status check

Microsoft SQL Server(DBaaS)

Performance item namedescriptionvalue
Instance State [Cluster]MSSQL cluster configuration statusPID: mssql when the process exists
-1: when the process does not exist
Instance State [PID]sqlservr.exe process pidFor Microsoft SQL Server, the secondary server also has a PID running, so the status cannot be determined solely by the PID.
Table. Microsoft SQL Server (DBaaS) status check

EPAS(DBaaS)

Performance item namedescriptionvalue
Instance State [PID]postgres process PIDPID: postgres if the process exists
* -1: if the process does not exist
Table. EPAS (DBaaS) status check

CacheStore(DBaaS)

Redis

Performance item nameExplanationvalue
Instance State [PID]Redis-server process PID-1: If the process does not exist
Sentinel State [PID]Sentinel process PID-1: when the process does not exist
Table. Redis status check

Valkey

Performance item namedescriptionvalue
Instance State [PID]Valkey-server process PID-1: If the process does not exist
Sentinel State [PID]Sentinel process PID-1: when the process does not exist
Table. Valkey status check

Data Analytics type

Event Streams

Performance item namedescriptionvalue
AKHQ State [PID]akhq process PIDPID: akhq if the process exists
* -1: if the process does not exist
Instance State [PID]Kafka process PIDPID: when the kafka process exists
* -1: when the process does not exist
Zookeeper State [Pid]zookeeper process PIDPID: zookeeper if the process exists
* -1: if the process does not exist
Table. Event Streams status check

Search Engine

Performance item nameExplanationvalue
Instance State [PID]Elasticsearch process PIDPID: if the Elasticsearch process exists
* -1: if the process does not exist
Kibana State [PID]Kibana process PIDPID: Kibana if the process exists
* -1: if the process does not exist
Table. Search Engine status check

Elasticsearch

Performance item nameExplanationvalue
Instance State [PID]Elasticsearch process PID-1: when the process does not exist
Kibana State [PID]Dashboard process PID-1: if the process does not exist
Table. Elasticsearch status check

Opensearch

Performance item namedescriptionvalue
Instance State [PID]Opensearch process PID-1: If the process does not exist
Dashboard State [PID]Dashboard process PID-1: when the process does not exist
Table. Opensearch status check

Vertica(DBaaS)

Performance item nameExplanationvalue
Instance State [PID]Vertica process PID-1: if the process does not exist
Table. Vertica (DBaaS) status check

Container type

Kubernetes Engine

Performance item namedescriptionvalue
Instance Statecluster status1: If the health check query sum(up{job=““kubernetes-apiservers””}) returns a value greater than 0
  • 0: If the health check query sum(up{job=““kubernetes-apiservers””}) returns a value less than or equal to 0 |
Table. Kubernetes Engine status check

Container Registry

Performance item namedescriptionvalue
Instance StateContainer Registry status1: running (normal)
* 0: down (abnormal)
Table. Container Registry status check

Networking type

Internet Gateway

Performance item namedescriptionvalue
N/AN/AN/A
Table. Internet Gateway status check
Caution
Internet Gateway does not provide status information through Cloud Monitoring.

Load Balancer(OLD)

Performance item namedescriptionvalue
Instance StateLoad Balancer statusDetermine based on provisioning_status in the API response
* 1: ACTIVE
* 0: ETC
Table. Load Balancer(OLD)

Load Balancer Listener(OLD)

Performance item nameExplanationvalue
Instance StateLoad Balancer Listener statusDetermine based on provisioning_status in the API response
* 1: ACTIVE
* 0: ETC
Table. Load Balancer Listener(OLD)

Load Balancer

Performance item namedescriptionvalue
Instance StateLoad Balancer statusDetermine based on provisioning_status in the API response
* 1: ACTIVE
* 0: ETC
Table. Load Balancer

Load Balancer Listener

Performance item namedescriptionvalue
Instance StateLoad Balancer Listener statusDetermine based on provisioning_status in the API response
* 1: ACTIVE
* 0: ETC
Table. Load Balancer Listener

Load Balancer Server Group

Performance item nameExplanationvalue
Instance StateStatus of Load Balancer Server GroupDetermine based on provisioning_status in the API response
* 1: ACTIVE
* 0: ETC
Table. Load Balancer Server Group

Direct Connect

Performance item namedescriptionvalue
N/AN/AN/A
Table. Direct Connect status check
Caution
Direct Connect does not provide status information through Cloud Monitoring.

Cloud WAN

Performance item namedescriptionvalue
Instance StateAttachment connection status0: down
* 1: up
* 2: testing
* 3: unknown
Table. Cloud WAN status check

Global CDN

Performance item namedescriptionvalue
Instance StateGlobal CDN status1: running (normal)
* 0: down (abnormal)
Table. Global CDN status check

3 - API Reference

API Reference

4 - Release Note

Cloud Monitoring

2025.07.01
FEATURE Add Cloud Monitoring integration service
  • In July 2025, we added an integrated service with Cloud Monitoring.
    • Additional integrated services: Compute(Multi-node GPU Cluster [Cluster Fabric], Multi-node GPU Clutser [Node]), Storage(Block Storage(BM), Block Storage(VM)), Networking(Cloud WAN, Global CDN), Database(Valkey), Data Analytics(Opensearch, Vertica(DBaaS))
2025.02.27
FEATURE Add Cloud Monitoring integration service
  • In February 2025, we added an integration service with Cloud Monitoring.
    • Additional integrated services: Container (Container Registry), Database (EPAS, Microsoft SQL Server), Data Analytics (Event Streams, Search Engine), Networking (Load Balancer, Load Balancer Listener, Load Balancer Server Group, VPN)
2024.10.01
NEW Official release of Cloud Monitoring service
  • We have launched the Cloud Monitoring service. It collects usage status and change information of operational infrastructure resources, and supports a stable cloud operating environment by generating and notifying events when configured thresholds are exceeded.