Managed Service for Prometheus seamlessly integrates with various Alibaba Cloud services, offering comprehensive metric data collection from cloud services.
The data plane for Managed Service for Prometheus offers two storage types: locally redundant storage and zone-redundant storage. These types provide data redundancy across a single zone or multiple zones to ensure data durability and availability. Managed Service for Prometheus is currently upgrading its storage redundancy features, and the availability of this feature varies by region. The upgrade status is consistent with that of Simple Log Service. For more information about supported regions, see Storage redundancy.
On the Managed Service for Prometheus data plane, locally redundant storage and zone-redundant storage have the same service availability and price.
Background information
Managed Service for Prometheus is capable of ingesting monitoring data from various Alibaba Cloud services and supports default Grafana dashboards and alerting rules.
Through Prometheus instances, you can uniformly collect, store, and display metrics.
Managed Service for Prometheus is natively integrated with other cloud services. After connecting the monitoring data of a cloud service via Integration Center or through default integrations, a corresponding Prometheus instance will be automatically created to store the related metric data.
Creation policy
Alibaba Cloud service | Description |
Container services, including Container Service for Kubernetes (ACK), Container Compute Service (ACS), ACK Serverless, and ACK One | A corresponding Prometheus instance will be created for each container cluster to store the metric data collected from the container service. |
Elastic Compute Service (ECS) | When integrating an ECS instance through Integration Center, a corresponding Prometheus instance will be created for the ECS instance and virtual private cloud (VPC). This Prometheus instance will store the metric data collected from the target ECS instance and all other ECS instances within the same VPC, consolidating the storage in the Prometheus instance. |
Realtime Compute for Apache Flink | If a Flink workspace is monitored in Managed Service for Prometheus, a corresponding Prometheus instance will be created for each workspace to store the relevant metric data collected from Flink. |
Other cloud services | After metrics from cloud services in the same region are integrated into Managed Service for Prometheus, a default Prometheus instance |
After creating an instance with Managed Service for Prometheus, it defaults to integrating with Managed Service for Grafana Shared Edition for convenient data querying. In Managed Service for Grafana Shared Edition, corresponding data sources will be automatically created, named in the format of prom-${PrometheusInstanceId}.
Storage upgrade
Starting 00:00:00 (UTC+8) on March 12, 2025, Managed Service for Prometheus gradually rolls out an upgrade to its storage infrastructure. This upgrade transitions metric data to be stored in the Metricstores of your Simple Log Service projects, enhancing performance and flexibility.
Change
Managed Service for Prometheus will automatically create Metricstores in your Simple Log Service projects to store metric data.
Here're the formats required:
Projects:
workspace-default-cms-{User ID}-{Region ID}
Metricstores:
ACK clusters: aliyun-prom-{Cluster ID}
Others: aliyun-prom-{Managed Service for Prometheus instance ID}
Impact
This upgrade has no impact on your historical data and incurs no additional cost.