Managed Service for Prometheus is a fully managed monitoring service on Alibaba Cloud that fully integrates with the open source Prometheus ecosystem, supporting a wide variety of component monitoring. It provides multiple out-of-the-box preconfigured dashboards and collects, stores, and analyzes metrics across your entire stack -- containers, hosts, applications, and cloud services.
The service is 100% compatible with the open source Prometheus ecosystem, including PromQL, Remote Write, Recording Rules, ServiceMonitor, and PodMonitor.
Prometheus instances
A Prometheus instance is the core resource unit in Managed Service for Prometheus. Each instance includes:
Data collection configuration -- defines which targets to scrape and which metrics to ingest.
Time-series database -- stores ingested metric data with configurable retention.
Dashboards -- preconfigured monitoring panels for common workloads.
Alert configuration -- alerting rules that evaluate metrics and generate events.
How it works
Managed Service for Prometheus provides a unified pipeline: collect metrics from any source, store them in a managed time-series database, then consume the data through dashboards, alerts, queries, and exports.
Data collection
Managed Service for Prometheus supports full-stack metric collection. The following table summarizes the supported sources and how each integrates with the service.
| Source | Integration method | Details |
|---|---|---|
| Client applications | Automatic via RUM | Real User Monitoring (RUM) metrics are written to Managed Service for Prometheus by default. |
| Server-side applications | Automatic via APM | Application Performance Monitoring (APM) metrics are written by default. For custom application metrics, use the Prometheus SDK or OpenTelemetry SDK. |
| Containers | Built-in Prometheus agent | Container Service for Kubernetes (ACK), ACK Serverless, and Container Compute Service (ACS) integrate by default. The self-developed Prometheus agent enables basic container metric collection. Define ServiceMonitor or PodMonitor resources for custom metric collection. |
| Self-managed Kubernetes | ACK One registration | Register external Kubernetes clusters (on-premises or other cloud providers) with ACK One. Once registered, metric collection works the same way as native ACK clusters. |
| Hosts (ECS) | Managed Prometheus agent | Collects CPU, memory, disk, network, and other OS-level metrics. Also supports collecting process- and container-related metrics. Custom metrics are supported through methods similar to the textfile format of node-exporter. |
| Hosts (other environments) | Remote Write | Deploy open source node-exporter and Prometheus in your data center or other cloud environment, then forward metrics to Managed Service for Prometheus via Remote Write. |
| Cloud services | CloudMonitor integration | Import cloud service metrics through the console. The integration automatically enriches metrics with instance names and tags for multi-dimensional aggregation, filtering, and alert routing. |
| Custom metrics | SDK | Ingest custom metrics through the Prometheus SDK or OpenTelemetry Metrics SDK. |
Data storage
Managed Service for Prometheus uses tiered storage to balance performance and cost:
Standard storage -- the primary storage tier for active metrics. Supports configurable retention periods. Billed based on data write volume or data reporting volume.
Archive storage -- for long-term retention at lower cost. When standard storage data exceeds the configured retention period, it is automatically moved to archive storage. Billed based on storage volume.
Data consumption
The following table lists the ways to query, visualize, and act on stored metrics.
| Capability | Description |
|---|---|
| Dashboards | Out-of-the-box dashboards for cloud services and common open source components. Import them into Managed Service for Grafana, or build custom dashboards. Integrate with Alibaba Cloud DataV through the Prometheus HTTP API for advanced visualization. |
| Global view | Query across multiple Prometheus instances and Alibaba Cloud accounts through the aggregated view. |
| Recording Rules | Pre-aggregate, down-sample, or reduce the dimensionality of metrics using Recording Rules (compatible with open source Recording Rule specifications) to lower query costs and improve performance. |
| Alerts | Built-in alerting rules for cloud services and common components. Create custom rules with PromQL or import rules from open source Prometheus. Alert events flow into Alert Management for notification, assignment, and escalation. |
| CloudLens | Unified observability across logs, metrics, and events for core monitoring scenarios. |
| PromQL analysis | 100% PromQL-compatible query engine. Explore metrics interactively, and use metric management to analyze metric distribution and high cardinality. |
| Data export | Stream metrics in real time to Kafka, MaxCompute, or a self-managed Prometheus instance. |
Benefits
Managed infrastructure:
A one-stop solution that simplifies the process of setting up and maintaining observability systems.
Tiered storage manages data lifecycle and optimizes costs without manual intervention.
Cost optimization:
Managed Service for Prometheus provides a more cost-effective and efficient solution compared to self-managed solutions.
Open source compatibility:
100% compatible with PromQL, Remote Write, Recording Rules, and the Prometheus data model.
Deep Alibaba Cloud integration:
Native integration with ACK, ACS, ECS, CloudMonitor, Managed Service for Grafana, and Alert Management.
Cloud service metric collection with automatic tag enrichment.
Enterprise readiness:
Cross-instance and cross-account aggregation for centralized monitoring.
Real-time data export to Kafka, MaxCompute, and external Prometheus instances.
Service-level agreement (SLA) guarantees and professional technical support.