horizontal pod autoscaling (HPA) in an Amazon Elastic Kubernetes Service (EKS)

Verify that the metrics-server is installed on your cluster. The metrics-server is a component that provides metrics for HPA to use. You can check if it is installed by running
kubectl get deployment metrics-server

Create an HPA resource for your application. You can create an HPA resource by creating a YAML file that defines the resource, and then applying it to your cluster using kubectl apply. An example HPA resource for an application named “myapp” might look like this:

apiVersion: autoscaling/v2beta2
kind: HorizontalPodAutoscaler
metadata:
  name: myapp-hpa
  namespace: default
spec:
  scaleTargetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: myapp
  minReplicas: 2
  maxReplicas: 10
  metrics:
  - type: Resource
    resource:
      name: cpu
      target:
        type: Utilization
        averageUtilization: 50

Verify that the HPA is working by checking its status. You can do this by running kubectl describe hpa myapp-hpa. The output should show the current and desired number of replicas, and the current CPU utilization.

You can test the HPA by increasing the load on the application, and then checking if the number of replicas increases.

Note

The above steps are general guideline and assumes that your cluster already have necessary permissions and configurations, please follow the EKS documentation for specific version and more details on the steps.
Horizontal Pod Autoscaling