AWS Certified DevOps Engineer – Professional (DOP-C02) — Question 380

A DevOps engineer is researching the least expensive way to implement an image batch processing cluster on AWS. The application cannot run in Docker containers and must run on Amazon EC2. The batch job stores checkpoint data on an NFS volume and can tolerate interruptions. Configuring the cluster software from a generic EC2 Linux image takes 30 minutes.

What is the MOST cost-effective solution?

Answer options

• A. Use Amazon EFS for checkpoint data. To complete the job, use an EC2 Auto Scaling group and an On-Demand pricing model to provision EC2 instances temporarily.
• B. Use GlusterFS on EC2 instances for checkpoint data. To run the batch job, configure EC2 instances manually. When the job completes, shut down the instances manually.
• C. Use Amazon EFS for checkpoint data. Use EC2 Fleet to launch EC2 Spot Instances, and utilize user data to configure the EC2 Linux instance on startup.
• D. Use Amazon EFS for checkpoint data. Use EC2 Fleet to launch EC2 Spot Instances. Create a custom AMI for the cluster and use the latest AMI when creating instances.

Correct answer: D

Explanation

Using EC2 Spot Instances via EC2 Fleet provides the most cost-effective compute option for workloads that can tolerate interruptions, while Amazon EFS provides the required NFS-compatible storage for checkpoint data. Since configuring the software from a generic image takes 30 minutes, using a pre-configured custom AMI (Option D) avoids this startup delay and allows instances to begin processing immediately, making it much more efficient than using user data (Option C). On-Demand instances (Option A) and manual EC2/GlusterFS configurations (Option B) are significantly more expensive and less automated.