Question #1983
A media streaming company is migrating from a monolithic architecture to a cloud-based solution. The solution must support scalable user growth, use modular microservices, maintain structured database schemas for user data, and provide cost-effective storage for high-resolution media files. Which option meets these requirements with the LEAST operational overhead?
Deploy a containerized application on Amazon EC2 instances in an Auto Scaling group. Use an Application Load Balancer for traffic distribution. Store user data and media files in an Amazon RDS Multi-AZ DB instance.
Refactor the application into AWS Lambda functions with Amazon API Gateway. Use Amazon DynamoDB for user data and media storage. Implement Amazon EventBridge for event-driven communication between services.
Use Amazon EKS with self-managed EC2 nodes to host a containerized application. Store user data in an Amazon Aurora Serverless cluster. Use AWS Batch for media processing and store media files in Amazon S3 Glacier Flexible Retrieval.
Deploy a containerized application using Amazon ECS with AWS Fargate. Store user data in an Amazon Aurora Multi-AZ database. Use Amazon S3 Standard-Infrequent Access for media file storage.
Explanation
Option D is correct because:
- Amazon ECS with AWS Fargate provides a serverless container platform, enabling scalable microservices with minimal operational overhead.
- Amazon Aurora Multi-AZ ensures a structured, highly available relational database for user data.
- Amazon S3 Standard-IA offers cost-effective storage for high-resolution media files.
Other options fail because:
- A: Uses EC2 (high operational overhead) and RDS for media (costly).
- B: DynamoDB (NoSQL) lacks structured schemas, and storing media in DynamoDB is expensive.
- C: Self-managed EKS nodes increase operational overhead; Glacier is unsuitable for frequent media access.
Key Points: Use managed services (Fargate, Aurora, S3) for scalability, structured data, and cost efficiency. Avoid EC2/self-managed services and NoSQL when structured schemas are required.
Answer
The correct answer is: D