Amazon Managed Service for Prometheus launches query insights and control capabilities

Amazon Managed Service for Prometheus, a fully managed Prometheus-compatible monitoring service, now provides the capability to identify expensive PromQL queries, and limit their execution. This enables customers to monitor and control the types of queries being issued against their Amazon Managed Service for Prometheus workspaces.

Customers have highlighted the need for tighter governance controls for queries, specifically around high cost queries. You can now monitor queries above a certain Query Samples Processed (QSP) threshold, and log those queries to Amazon CloudWatch. The information in the vended logs allows you to identify expensive queries. The vended logs contain the PromQL query and metadata about where it originated from, such as from Grafana dashboard IDs or alerting rules.
In addition, you can now set warning or error thresholds for query execution. To control query cost, you can pre-empt the execution of expensive queries by providing an error threshold in the HTTP headers to the QueryMetrics API. Alternatively, by setting a warning threshold, we return the query results, charge you for the QSP, and return a warning to the end-user that the query is more expensive than the limit set by your workspace administrator.

This feature is now available in all regions where Amazon Managed Service for Prometheus is generally available.

To learn more about Amazon Managed Service for Prometheus collector, visit the user guide or product page.