AWS Certified Solutions Architect – Associate (SAA-C03) — Question 683

A company runs container applications by using Amazon Elastic Kubernetes Service (Amazon EKS) and the Kubernetes Horizontal Pod Autoscaler. The workload is not consistent throughout the day. A solutions architect notices that the number of nodes does not automatically scale out when the existing nodes have reached maximum capacity in the cluster, which causes performance issues.

Which solution will resolve this issue with the LEAST administrative overhead?

Answer options

• A. Scale out the nodes by tracking the memory usage.
• B. Use the Kubernetes Cluster Autoscaler to manage the number of nodes in the cluster.
• C. Use an AWS Lambda function to resize the EKS cluster automatically.
• D. Use an Amazon EC2 Auto Scaling group to distribute the workload.

Correct answer: B

Explanation

The Kubernetes Cluster Autoscaler is the standard, native solution to automatically adjust the number of nodes in an Amazon EKS cluster when pods cannot be scheduled due to resource constraints, minimizing administrative overhead. While Amazon EC2 Auto Scaling groups manage the underlying instances, they do not natively understand Kubernetes pod scheduling requirements without the Cluster Autoscaler or Karpenter. Custom solutions like AWS Lambda functions or scaling based purely on memory metrics introduce unnecessary complexity and operational overhead.