Container Service for Kubernetes:Deploy AHPA for predictive autoscaling

Step 1: Install AHPA Controller

1. Log on to the ACK console. In the left navigation pane, click Clusters.

2. On the Clusters page, click the name of your cluster. In the left navigation pane, click Components and Add-ons .

3. On the Add-ons page, locate AHPA Controller and click Install. Follow the prompts.

Step 2: Configure Prometheus data source

1. Log on to the ARMS console.

2. In the left navigation pane, choose Managed Service for Prometheus > Instances.

3. On the Instances page, select the region, then click the instance named after your ACK cluster.

4. On the Settings page, in the HTTP API Address (Grafana Read Address) section, record:

   • (Optional) The access token, if token-based authentication is enabled.

   • The Internal Network endpoint (Prometheus URL).

5. Set the Prometheus query URL in your ACK cluster.

   1. Create application-intelligence.yaml with the following content.

      • prometheusUrl: The Prometheus endpoint.

      • token: The access token for Prometheus.

      apiVersion: v1
      kind: ConfigMap
      metadata:
        name: application-intelligence
        namespace: kube-system
      data:
        prometheusUrl: "http://cn-hangzhou-intranet.arms.aliyuncs.com:9443/api/v1/prometheus/da9d7dece901db4c9fc7f5b9c40****/158120454317****/cc6df477a982145d986e3f79c985a****/cn-hangzhou"
        token: "eyJhxxxxx"

      Note

      To view the AHPA dashboard in Prometheus, also configure these fields in this ConfigMap:

      • prometheus_writer_url: The internal endpoint for Remote Write.

      • prometheus_writer_ak: The AccessKey ID of your Alibaba Cloud account.

      • prometheus_writer_sk: The AccessKey secret of your Alibaba Cloud account.

      See Enable the Prometheus dashboard for AHPA.

   2. Deploy application-intelligence.

      kubectl apply -f application-intelligence.yaml

Step 3: Deploy a test service

The test service includes a Deployment fib-deployment, a Service fib-svc, and a load generator fib-loader that simulates traffic. An HPA provides a baseline for comparison with AHPA.

apiVersion: apps/v1
kind: Deployment
metadata:
  name: fib-deployment
  namespace: default
  annotations:
    k8s.aliyun.com/eci-use-specs: "1-2Gi"
spec:
  replicas: 1
  selector:
    matchLabels:
      app: fib-deployment
  strategy:
    rollingUpdate:
      maxSurge: 25%
      maxUnavailable: 25%
    type: RollingUpdate
  template:
    metadata:
      creationTimestamp: null
      labels:
        app: fib-deployment
    spec:
      containers:
      - image: registry.cn-huhehaote.aliyuncs.com/kubeway/knative-sample-fib-server:20200820-171837
        imagePullPolicy: IfNotPresent
        name: user-container
        ports:
        - containerPort: 8080
          name: user-port
          protocol: TCP
        resources:
          limits:
            cpu: "1"
            memory: 2000Mi
          requests:
            cpu: "1"
            memory: 2000Mi
---
apiVersion: v1
kind: Service
metadata:
  name: fib-svc
  namespace: default
spec:
  ports:
  - name: http
    port: 80
    protocol: TCP
    targetPort: 8080
  selector:
    app: fib-deployment
  sessionAffinity: None
  type: ClusterIP
---
apiVersion: apps/v1
kind: Deployment
metadata:
  name: fib-loader
  namespace: default
spec:
  progressDeadlineSeconds: 600
  replicas: 1
  revisionHistoryLimit: 10
  selector:
    matchLabels:
      app: fib-loader
  strategy:
    rollingUpdate:
      maxSurge: 25%
      maxUnavailable: 25%
    type: RollingUpdate
  template:
    metadata:
      creationTimestamp: null
      labels:
        app: fib-loader
    spec:
      containers:
      - args:
        - -c
        - |
          /ko-app/fib-loader --service-url="http://fib-svc.${NAMESPACE}?size=35&interval=0" --save-path=/tmp/fib-loader-chart.html
        command:
        - sh
        env:
        - name: NAMESPACE
          valueFrom:
            fieldRef:
              apiVersion: v1
              fieldPath: metadata.namespace
        image: registry.cn-huhehaote.aliyuncs.com/kubeway/knative-sample-fib-loader:20201126-110434
        imagePullPolicy: IfNotPresent
        name: loader
        ports:
        - containerPort: 8090
          name: chart
          protocol: TCP
        resources:
          limits:
            cpu: "8"
            memory: 16000Mi
          requests:
            cpu: "2"
            memory: 4000Mi
---
apiVersion: autoscaling/v1
kind: HorizontalPodAutoscaler
metadata:
  name: fib-hpa
  namespace: default
spec:
  maxReplicas: 50
  minReplicas: 1
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: fib-deployment
  targetCPUUtilizationPercentage: 50
---

Step 4: Deploy AHPA

To configure a scaling policy, create an AdvancedHorizontalPodAutoscaler resource.

1. Create ahpa-demo.yaml with the following content.

   apiVersion: autoscaling.alibabacloud.com/v1beta1
   kind: AdvancedHorizontalPodAutoscaler
   metadata:
     name: ahpa-demo
   spec:
     scaleStrategy: observer
     metrics:
     - type: Resource
       resource:
         name: cpu
         target:
           type: Utilization
           averageUtilization: 40
     scaleTargetRef:
       apiVersion: apps/v1
       kind: Deployment
       name: fib-deployment 
     maxReplicas: 100
     minReplicas: 2
     stabilizationWindowSeconds: 300
     prediction:
       quantile: 95
       scaleUpForward: 180
     instanceBounds:
     - startTime: "2021-12-16 00:00:00"
       endTime: "2031-12-16 00:00:00"
       bounds:
       - cron: "* 0-8 ? * MON-FRI"
         maxReplicas: 15
         minReplicas: 4
       - cron: "* 9-15 ? * MON-FRI"
         maxReplicas: 15
         minReplicas: 10
       - cron: "* 16-23 ? * MON-FRI"
         maxReplicas: 20
         minReplicas: 15

   Selected parameters.

   Parameter	Required	Description
   scaleTargetRef	Yes	The target Deployment.
   metrics	Yes	The scaling metrics. Supported metrics include CPU, GPU, memory, QPS, and RT.
   target	Yes	The target threshold. For example, averageUtilization: 40 targets 40% CPU utilization.
   scaleStrategy	No	The scaling mode. Default: observer. auto: AHPA performs scaling. observer: AHPA observes without scaling. Use to verify AHPA behavior. proactive: Only proactive prediction takes effect. reactive: Only reactive scaling takes effect.
   maxReplicas	Yes	The maximum replica count.
   minReplicas	Yes	The minimum replica count.
   stabilizationWindowSeconds	No	The stabilization window to prevent rapid replica fluctuations. Default: 300 seconds.
   prediction.quantile	Yes	The prediction quantile (0–100). Higher values yield more conservative predictions. Recommended: 90–99. Default: 99.
   prediction.scaleUpForward	Yes	Time for a new pod to become Ready, also known as the cold start time.
   instanceBounds	No	Replica count boundaries for specific time periods. startTime: The start time. endTime: The end time.
   instanceBounds.bounds.cron	No	A cron expression for scheduled tasks. For example, - cron: "* 0-8 ? * MON-FRI" runs every minute from 00:00 to 08:59, Monday to Friday.

   Cron expression fields. See Cron scheduled tasks.

   Field	Required	Allowed values	Special characters
   Minutes	Yes	0–59	* / , -
   Hours	Yes	0–23	* / , -
   Day of month	Yes	1–31	* / , - ?
   Month	Yes	1–12 or JAN–DEC	* / , -
   Day of week	Yes	0–6 or SUN–SAT	* / , - ?

   Note

   • Month and Day of week are case-insensitive. For example, SUN, Sun, and sun are equivalent.

   • Day of week defaults to * if not specified.

   • Special characters:

     • *: Matches all values.

     • /: Specifies an increment.

     • ,: Lists enumerated values.

     • -: Specifies a range.

     • ?: No specific value. Use in Day of month or Day of week when the other is set.

2. Create the AHPA scaling policy.

   kubectl apply -f ahpa-demo.yaml

Step 5: View prediction results

Enable the Prometheus dashboard for AHPA to view prediction results.

This example uses the observer scaling mode (observer mode) to compare AHPA predictions against an HPA baseline for actual resource needs.

Related documents

• See Configure AHPA predictive scaling based on GPU metrics.

• See Enable the Prometheus dashboard for AHPA.