Managed Service for Prometheus tracks resource consumption across your instances in two dimensions: reporting volume (measured in data points) and writing volume (measured in bytes). Use the Usage Statistics page to monitor this data and manage costs.
View usage data
In the left-side navigation pane, click Usage Statistics.
In the upper-right corner, set the time range to view usage data for a specific period.
Usage data retention
To meet custom statistical requirements, starting November 25, 2024, you can use Managed Service for Prometheus to create Simple Log Service Metricstores for Prometheus usage data. The retention duration is 180 days. You can run custom queries on this historical data to analyze usage trends and identify cost optimization opportunities.
The free tier includes:
Storage capacity
Indexing
Read/write traffic
Read/write operations
Note: Other Simple Log Service features may incur fees.
Metricstore settings
| Setting | Value |
|---|---|
| Project | aliyun-product-data-{userId}-cn-shanghai |
| Metricstore | prom_finance_aggr |
Metrics
Managed Service for Prometheus records usage in two dimensions:
Reporting volume (points): The number of data points reported by your monitoring targets. This reflects how many metric samples are collected per scrape interval.
Writing volume (bytes): The amount of data written to storage. This reflects the actual data size of ingested metrics.
Each dimension separates basic metrics (free tier) from custom metrics (billed).
| Metric | Dimension | Tier | Unit | Description |
|---|---|---|---|---|
prom_points_job_total_cost | Reporting volume | Custom (billed) | Points | Data points reported by custom metrics at the job level |
prom_points_job_total_free | Reporting volume | Basic (free) | Points | Data points reported by basic metrics at the job level |
prom_serverless_job_amount_cost | Writing volume | Custom (billed) | Bytes | Data written by custom metrics at the job level |
prom_serverless_job_amount_free | Writing volume | Basic (free) | Bytes | Data written by basic metrics at the job level |
prom_serverless_archive_amount | Writing volume | Archival | Bytes | Data billed by writing volume that enters archival storage |
Tags
Use these tags to filter and aggregate usage data by instance, region, or job:
| Tag | Description |
|---|---|
instanceId | Managed Service for Prometheus instance ID |
instanceName | Managed Service for Prometheus instance name |
regionId | Region where the instance is deployed |
job | Sampling task name |
userId | Alibaba Cloud account ID |
Reduce costs
To lower your Managed Service for Prometheus costs, reduce the number of custom metrics you report. Custom metrics (_cost suffix) are billed, while basic metrics (_free suffix) are free.
Discard unused metrics: If you no longer need specific metrics, . Discarded metrics are not billed.
Re-enable metrics: To resume monitoring previously discarded metrics, . Standard billing applies after re-enablement.