AWS Certified Solutions Architect - Associate / Question #1975 of 1019

Question #1975

A company is migrating its analytics platform to AWS. The platform executes short-lived batch jobs that are fault-tolerant. Data is produced after each batch job and is actively accessed for 45 days before being archived for compliance for 5 years. The company wants to minimize costs without affecting the batch job processing. Which solution meets these requirements?

A

Use Amazon EC2 Spot Instances for the batch jobs. Store data in Amazon S3 Standard-Infrequent Access. Transition to Amazon S3 Glacier Flexible Retrieval after 45 days. Set a lifecycle policy to delete data after 5 years.

B

Deploy Amazon EC2 On-Demand Instances for the batch jobs. Store data in Amazon S3 Standard. Transition to Amazon S3 Glacier Deep Archive after 45 days. Configure expiration to delete data after 5 years.

C

Utilize Amazon EC2 Spot Instances to run the batch jobs. Store data in Amazon S3 Standard. Transition to Amazon S3 Glacier Deep Archive after 45 days. Set an expiration policy to delete the data after 5 years.

D

Migrate the batch jobs to Amazon EC2 Reserved Instances. Store data in Amazon S3 Intelligent-Tiering. Move data to Amazon S3 Glacier Deep Archive after 45 days. Expire data after 5 years.

Explanation

Option C is correct because:
- Spot Instances minimize compute costs for fault-tolerant, short-lived batch jobs.
- S3 Standard is optimal for data actively accessed for 45 days (Standard-IA in Option A incurs retrieval costs and minimum storage fees).
- Glacier Deep Archive is the cheapest storage for 5-year compliance archiving (vs. Glacier Flexible Retrieval in Option A).
- Expiration policy deletes data after 5 years.

Other options are incorrect:
- Option A: Uses Standard-IA (higher costs for frequent access) and Glacier Flexible Retrieval (more expensive than Deep Archive).
- Option B: Uses On-Demand Instances (higher compute costs than Spot).
- Option D: Reserved Instances are less flexible for short jobs, and Intelligent-Tiering adds unnecessary monitoring fees.

Key Points:
- Use Spot Instances for fault-tolerant, short jobs.
- S3 Standard for frequent access; transition to Glacier Deep Archive for long-term storage.
- Lifecycle policies automate cost-optimized storage transitions.

Answer

The correct answer is: C