Container Compute Service:Horizontal Pod Autoscaler (HPA)

Use the console

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

2. On the Clusters page, click the name of the target cluster. In the left navigation pane, choose Workloads > Deployments.

3. On the Deployments page, click Create with Image.

4. On the Basic Information page, set the application parameters and click Next.

   Parameter	Description
   Namespace	The namespace for the application. Defaults to default.
   Application Name	The name of the application.
   Replicas	The number of pods for the application. Default: 2.
   Workload	The type of workload. Select Deployments, StatefulSets, Jobs, or CronJobs.
   Labels	A label to identify the application.
   Annotations	An annotation for the application.
   Instance Type	The instance type of the pod. Select General-purpose, BestEffort, or Performance-enhanced.

5. On the Container page, configure the container, select an image, and specify resource requests. Then, click Next. For more information, see Container configuration.

   Note

   HPA requires resource requests to be specified for the Deployment.

6. On the Advanced page, in the Access Control section, click Create next to Service, and configure the Service parameters. For more information, see Advanced configurations.

7. On the Advanced page, select Enable for HPA and configure the scaling parameters.

   • Metric: The supported metrics are CPU and memory. Must match the configured required resource type.

   • Condition:: The resource utilization percentage. The container scales out when utilization exceeds this value. For more information about the Horizontal Pod Autoscaling algorithm, see Algorithm details.

   • Max. Replicas: The maximum number of containers the Deployment can run.

   • Min. Replicas: The lower limit on the number of replicas.

8. Click Create to create the HPA-enabled Deployment.

Use kubectl

You can also use an orchestration template and kubectl to create an HPA object and bind it to a Deployment.

The following example uses an NGINX application.

1. Create a file named nginx.yaml and copy the following content into it.

   apiVersion: apps/v1 
   kind: Deployment
   metadata:
     name: nginx
     labels:
       app: nginx
   spec:
     replicas: 2
     selector:
       matchLabels:
         app: nginx  
     template:
       metadata:
         labels:
           app: nginx
       spec:
         containers:
         - name: nginx
           image: registry-cn-hangzhou.ack.aliyuncs.com/ack-demo/nginx:1.7.9 # Replace with your image name and tag
           ports:
           - containerPort: 80
           resources:
             requests:                         ## Required for HPA to function.
               cpu: 500m

2. Run the following command to deploy the NGINX application.

   kubectl apply -f nginx.yaml

3. Create an hpa.yaml file with the following content to define the HPA.

   The scaleTargetRef field specifies the resource that HPA scales. In this example, it targets the nginx Deployment.

   apiVersion: autoscaling/v2
   kind: HorizontalPodAutoscaler
   metadata:
     name: nginx-hpa
   spec:
     scaleTargetRef:
       apiVersion: apps/v1
       kind: Deployment
       name: nginx
     minReplicas: 1  # The minimum replica count. Must be >= 1.
     maxReplicas: 10 # The maximum replica count. Must be > minReplicas.
     metrics:
     - type: Resource
       resource:
         name: cpu
         target:
           type: Utilization
           averageUtilization: 50 # The target average utilization across all pods, calculated as (average usage / requested amount).
                  

4. Run the following command to create the HPA.

   kubectl apply -f hpa.yaml

5. After the HPA is created, run kubectl describe hpa <HPA_NAME>.

   The following output indicates that HPA is working correctly.

    Type    Reason             Age    From                       Message
     ----    ------             ----   ----                       -------
     Normal  SuccessfulRescale  4m53s  horizontal-pod-autoscaler  New size: 1; reason: All metrics below target