Amazon Managed Service for Prometheus adds out-of-order ingestion and rule query offset
Amazon Managed Service for Prometheus now supports out-of-order sample ingestion and a workspace-level rule query offset. These features help reduce data loss and improve alerting accuracy for workloads with distributed collectors or variable network latency. This update is available in all regions where the service is generally available and benefits users with intermittent ingestion delays.
- →Support for out-of-order sample ingestion
- →Workspace-level rule query offset
- →Visibility into ingestion patterns with new CloudWatch metrics
- →Feature availability and configuration
Features (2) ›
- Support for out-of-order sample ingestion
Workspaces now have a default 1-minute out-of-order time window, allowing ingestion of metric samples that arrive outside strict chronological order. This window can be adjusted or set to 0 to disable the feature.
- Workspace-level rule query offset
A global rule query offset can be configured to delay rule evaluation queries, ensuring late-arriving samples are ingested before rules execute. This compensates for expected ingestion delays and improves rule accuracy.
Enhancements (1) ›
- Visibility into ingestion patterns with new CloudWatch metrics
Two new CloudWatch vended metrics, OutOfOrderIngestionRate and OutOfOrderSampleAge, are now available. These metrics provide insight into ingestion patterns, aiding in the tuning of out-of-order settings.
Notes (1) ›
- Feature availability and configuration
Out-of-order sample ingestion and rule query offset are available in all AWS regions where Amazon Managed Service for Prometheus is generally available. Configuration can be done via the AWS console, API, or CLI.
https://aws.amazon.com/about-aws/whats-new/2026/06/amazon-managed-service-prometheus-outoforder-ingestion/