Question #1696
A company is setting up a staging environment on Amazon Elastic Kubernetes Service (Amazon EKS) to validate application updates. The staging cluster will be used intermittently to simulate failure scenarios and must minimize costs. The EKS cluster is required to fully manage all node lifecycle operations.
Which solution meets these requirements MOST cost-effectively?
Create a managed node group configured exclusively with Spot Instances.
Create two managed node groups: one with On-Demand Instances and another with Spot Instances.
Deploy an Auto Scaling group using Spot Instances and customize the user data to register nodes with the EKS cluster.
Create a managed node group that uses only On-Demand Instances.
Explanation
The correct answer is A. Managed node groups automate node lifecycle management, meeting the requirement for EKS to handle these operations. Spot Instances are the most cost-effective option for intermittent workloads, such as a staging environment used for failure simulations. While Spot Instances can be interrupted, this is acceptable in non-production environments.
Option B introduces On-Demand Instances, which are more expensive and unnecessary for intermittent use. Option C requires manual node registration via user data, violating the requirement for EKS-managed nodes. Option D uses only On-Demand Instances, which is not cost-optimized. Key points: Use managed node groups for lifecycle automation and Spot Instances for cost savings in non-critical environments.
Answer
The correct answer is: A