Prometheus Service allows you to view resource usage and learn about the data samples reported for the metrics of Prometheus instances. This topic describes how to view resource usage.
Procedure
- Log on to the Managed Service for Prometheus console.
- In the left-side navigation pane, click Monitoring List.
- On the Prometheus Service page, click Resource Usage.
- View resource usage on the Resource Usage page. Prometheus Service provides the following dashboards to show the resource usage of Prometheus instances.
- Resource Consumption: displays the overall monitoring data, including the data samples reported for the custom metrics of Prometheus instances in the current region. For more information, see Resource Consumption dashboard.
- Resource Consumption Cluster: displays monitoring data by cluster and provides the data sample distribution of each job for a specified Prometheus instance. For more information, see Resource Consumption Cluster dashboard.
- Resource Consumption Job: displays monitoring data by job, and provides the data sample distribution of custom metrics for a specified Prometheus instance. For more information, see Resource Consumption Job dashboard.
Note The metrics are classified into two types: basic metrics and custom metrics. For more information, see Basic metrics. Custom metrics refer to non-basic metrics.
Resource Consumption dashboard
In the upper-left corner of the dashboard, you can query monitoring data by TopK, Statistical interval, and instance.
- TopK: If you configure the filter condition, the top K Prometheus instances that have the most data samples reported for custom metrics, the corresponding data collection jobs, and the data samples are displayed. For example, if Topk is set to 20, you can view the top 20 Prometheus instances that have the most data samples reported for custom metrics, the corresponding data collection jobs, and the data samples.
- Statistical interval: the interval at which monitoring data is collected. If you set the value to 1h, monitoring data is collected and visualized once an hour.
- instance: You can view the monitoring data of a specific Prometheus instance.
The following table describes the items of the Resource Consumption dashboard.
| Section | Item | Description |
|---|---|---|
| 1 | Custom indicator report volume | These sections display the number of data samples reported for the custom metrics and basic metrics of the Prometheus instance within the last 24 hours. These sections also display the total number of custom metrics and basic metrics of the Prometheus instance. You can also configure a time period to view the total number of data samples reported within the time period. For example, if you set the statistical interval to 7 days, the Cumulative reported volume parameter displays the total number of data samples reported for custom metrics and basic metrics within the last seven days. |
| Basic Indicator Reported Volume | ||
| Number of custom indicators | ||
| Number of basic indicators | ||
| 2 | Custom indicator reporting trend | This section displays the trends in data reporting of the top K Prometheus instances that have the most data samples reported for custom metrics within a specified time period. Each point in the line chart indicates the total number of data samples reported for the custom metrics of the current Prometheus instance within a time period. |
| 3 | Instance details | This section displays the details of the current Prometheus instance, including the instance type, billing method, and number of data samples reported for metrics within the last 24 hours. If your Prometheus instance uses the subscription billing method, this section also displays metrics such as the throttling status, total number of reported data samples, and the upper limit of data samples that you can report. Important If the Prometheus instance is in the Throttled state, you need to check whether the instance has exceeded the upper limit of data samples that you can report for metrics. If the instance has exceeded the upper limit, you must upgrade the instance type to prevent your service from being affected. You can click the name of the Prometheus instance to go to the Resource Consumption Cluster dashboard. Then, you can view the resource usage of the Prometheus instance. |
| 4 | Custom indicator report (Job) | These sections display the jobs and custom metrics of the top K Prometheus instances that have the most data samples reported for custom metrics. You can locate abnormal data to troubleshoot issues.
|
| Custom indicator report (indicator) |
Resource Consumption Cluster dashboard
In the upper-left corner of the dashboard, you can query the monitoring data by TopK, Statistical interval, instance, and Job.
- TopK: If you configure the filter condition, the top K Prometheus instances that have the most data samples reported for custom metrics, the corresponding data collection jobs, and the data samples are displayed. For example, if Topk is set to 20, you can view the top 20 Prometheus instances that have the most data samples reported for custom metrics, the corresponding data collection jobs, and the data samples.
- Statistical interval: the interval at which monitoring data is collected. If you set the value to 1h, monitoring data is collected and visualized once an hour.
- Instance: You can view the monitoring data of a specific Prometheus instance.
- Job: You can view the monitoring data of a specific data collection job.
The following table describes the items of the Resource Consumption Cluster dashboard.
| Section | Item | Description | |
|---|---|---|---|
| 1 | Custom indicator report volume | 24-hour report | This section displays the number of data samples reported for the custom metrics of the current Prometheus instance within the last 24 hours. |
| Cumulative reported volume | This section displays the number of data samples reported for the custom metrics of the current Prometheus instance within a specified time period. For example, if you select Last 7 days in the upper-right corner, the Cumulative reported volume parameter displays the total number of data samples reported for custom metrics within the last seven days. | ||
| Type of payment | This section displays the billing methods of the top K Prometheus instances that have the most data samples reported for custom metrics. For more information about the billing methods provided by Prometheus Service, see Billing overview. Note If your Prometheus instance uses the subscription billing method, this section also displays metrics such as the total number of reported data samples and the upper limit of data samples that you can report. | ||
| Monthly Usage Reported Quantity | This section displays the number of data samples reported for the custom metrics of the subscription Prometheus instance. For more information about the subscription billing method, see Overview. | ||
| Monthly reporting limit | This section displays the upper limit of data samples that you can report for the custom metrics of the subscription Prometheus instance. For more information about the subscription billing method, see Overview. | ||
| 2 | Custom indicator reporting trend | This section displays the data collection jobs and the trends in data reporting of the top K Prometheus instances that have the most data samples reported for custom metrics within a specified time period. Each point in the line chart indicates the corresponding data collection job and the total number of data samples reported for the custom metrics of the current Prometheus instance within a time period. | |
| 3 | Custom Indicator Report (Job) | These sections display the jobs and custom metrics of the top K Prometheus instances that have the most data samples reported for custom metrics. You can locate abnormal data to troubleshoot issues. | |
| User-defined Indicator Reporting Quantity (Indicator) | |||
Resource Consumption Job dashboard
In the upper-left corner of the dashboard, you can query the monitoring data by TopK, Statistical interval, instance, Job, and Indicators.
- TopK: If you configure the filter condition, the top K Prometheus instances that have the most data samples reported for custom metrics, the corresponding data collection jobs, and the data samples are displayed. For example, if Topk is set to 20, you can view the top 20 Prometheus instances that have the most data samples reported for custom metrics, the corresponding data collection jobs, and the data samples.
- Statistical interval: the interval at which monitoring data is collected. If you set the value to 1h, monitoring data is collected and visualized once an hour.
- Instance: You can view the monitoring data of a specific Prometheus instance.
- Job: You can view the monitoring data of a specific data collection job.
- Metric: You can view the monitoring data of a specific metric.
The following table describes the items of the Resource Consumption Job dashboard.
| Section | Item | Description |
|---|---|---|
| 1 | Custom Indicator Reporting Trend | This section displays the data collection jobs and the trends in data reporting of the top K Prometheus instances that have the most data samples reported for custom metrics within a specified time period. Each point in the line chart indicates the corresponding data collection job and the total number of data samples reported for the custom metrics of the current Prometheus instance within a time period. |
| 2 | User-defined indicator reporting details | This section displays the jobs and custom metrics of the top K Prometheus instances that have the most data samples reported for custom metrics. You can locate abnormal data to troubleshoot issues. |