All Products
Search

This Product

    Cloud Monitor:CloudLens for ECS

    Document Center

    Cloud Monitor:CloudLens for ECS

    Last Updated:Jun 02, 2026

    CloudLens for ECS is a Cloud Monitor 2.0 application that provides comprehensive monitoring of host performance, processes, resources, and events.

    Getting started

    1. Log on to the Cloud Monitor 2.0 console. Select a target workspace and go to Integration Center. Choose an integration option for ECS observability:

      • Elastic Compute Service (ECS): Integrates ECS instance metrics, audit logs, and system events in one operation. Supports batch integration across an entire region regardless of VPC. Collected metrics include Cloud Monitor agentless metrics and Exporter proxy metrics. The system auto-detects Linux and Windows instances, connects to them, and installs the Exporter.

      • GPU Monitoring for ECS: Installs the Gpu-exporter on GPU instances to collect metrics. Supports Alibaba Cloud Linux, CentOS, and Ubuntu.

    2. Keep the default settings on the configuration page and click Determine.

    3. After integration, check the status in Integration Management under Integration Center. The system automatically creates a Simple Log Service (SLS) project to store collected metrics and logs.

    Application overview

    Log on to the Cloud Monitor 2.0 console. Select a target workspace, go to Application Center, and click CloudLens for ECS to open the HostOverview page. This page summarizes all ECS instances: total count, vCPUs, memory, health status, and observability coverage (the number of integrated instances).

    • Account-level instance analysis:

      • View instance statistics by account: total instances, requested vCPUs, and requested memory.

      • View instance statistics by region: total instances, requested vCPUs, and requested memory.

    • Account-level instance status analysis:

      • Analyze instance health with point-in-time comparisons, time-series analysis, and normal/abnormal counts.

      • Analyze GuestOS-caused abnormalities with point-in-time comparisons, time-series analysis, and normal/abnormal counts. An abnormal GuestOS status indicates a guest OS issue.

      • Analyze host-caused abnormalities with point-in-time comparisons, time-series analysis, and normal/abnormal counts. An abnormal host status indicates an underlying Alibaba Cloud physical host issue.

    • Observability coverage (agentless metrics):

      • View the count and percentage of instances using Cloud Monitor agentless metrics, their collection status, and the instance list.

    • Observability coverage (Exporter proxy metrics):

      • View the count and percentage of instances using Exporter proxy metrics, their collection status, and the instance list.

    • Resource analysis for integrated instances:

      • View resource statistics for integrated instances: vCPU count, requested and remaining memory and disk space, instance type distribution, and process/thread counts.

    Next steps

    • Resource Center: View your ECS instances and the processes running on them.

    • Insights: Monitor host resources, performance, processes, and events.

    • Alert Management: Configure alert rules and review your alert history.

    • Data Exploration: Query and analyze your metrics, logs, and events.