Container Service for Kubernetes:Enable the Prometheus dashboard for AHPA

Prerequisites

Before you begin, make sure that:

• Managed Service for Prometheus is enabled for your cluster and the Prometheus Agent is version 3.0.0 or later. For more information, see Enable Prometheus monitoring.

• AHPA is deployed and is version v2.0.0 or later. For more information, see Deploy AHPA.

• The application-intelligence ConfigMap in the kube-system namespace includes the following parameters:

  apiVersion: v1
  kind: ConfigMap
  metadata:
    name: application-intelligence
    namespace: kube-system
  data:
    prometheus_writer_url: http://cn-hangzhou-intranet.arms.aliyuncs.com/prometheus/190601496873****/cd5ec926b715e41b282296d1415a1***/cn-hangzhou/api/v3/write
    prometheus_writer_ak: xxx
    prometheus_writer_sk: xx
    ...

  Parameter	Description
  prometheus_writer_url	The internal endpoint for Remote Write
  prometheus_writer_ak	Your AccessKey ID
  prometheus_writer_sk	Your AccessKey Secret

Connect AHPA to Managed Service for Prometheus

1. Log on to the ARMS console.

2. In the left navigation pane, click Integration Center and search for ACK AHPA.

3. Click the ACK AHPA resource card, then follow the on-screen instructions to select your Container Service for Kubernetes (ACK) cluster and configure the connection.

   Parameter	Description
   Access Name	A unique name for this AHPA monitoring instance
   metrics collection interval (seconds)	The interval at which monitoring data is collected

View the AHPA dashboard

1. Log on to the ARMS console.

2. In the left navigation pane, click Integration Management, then click the Integrated Addons tab.

3. Click the ACK AHPA resource card, go to the Dashboards tab, and click a dashboard name to open it in Grafana.

The dashboard displays the following panels:

• CPU Utilization & Actual POD: The average CPU utilization and the current number of pods for the workload.

• Actual and Predicted CPU Usage: The total CPU usage of all pods alongside the predicted usage. When predicted usage exceeds actual usage, AHPA has reserved enough CPU capacity for the anticipated load.

• Pod Trends: The number of pods across three dimensions:

  • Actual Pods: The number of pods currently running.

  • Recommended Pods: The final pod count that AHPA recommends, based on proactive prediction, reactive prediction, and the boundary range.

  • Proactive Prediction: The pod count predicted from historical data and detected traffic periodicity.

Key AHPA metrics

The following metrics are available in Managed Service for Prometheus after you connect AHPA.

What's next

To scale your application based on custom metrics such as HTTP requests per second (QPS) or message queue length, see Use AHPA to configure custom metrics for application autoscaling.