New Release SAA-C03 AWS Certified Associate Questions

Amazon Kinesis is the right streaming service because it ingests and processes data in real time. For the user-data store, Amazon DynamoDB is the correct database because AWS documents that DynamoDB is a serverless, fully managed, distributed NoSQL database with single-digit millisecond performance at any scale. That directly matches the durability, low-latency, and unknown-scale requirements in the question. RDS is not the most natural fit for unpredictable massive serverless scaling here, while CloudFront and Global Accelerator solve delivery and network-edge problems rather than real-time event ingestion plus low-latency persistence. For a serverless analytics application with streaming game data and highly scalable user storage, Kinesis and DynamoDB are the best pair.

============

A. API Gateway + DynamoDB:Provides high scalability, low latency, and seamless integration with Amazon Comprehend for NLP tasks.

B. HTTP API + Translate + ElastiCache:Translate is not relevant for NLP tasks like sentiment analysis or entity recognition. ElastiCache is unsuitable for permanent storage.

C. SQS + EC2 + RDS:Increases complexity and operational overhead. RDS may not scale effectively for high concurrent loads.

D. WebSocket API + Textract:Textract is irrelevant for NLP tasks. WebSocket API is not the optimal choice for this use case.

Amazon DynamoDB supports two read consistency models: eventually consistent reads and strongly consistent reads. By default, DynamoDB uses eventual consistency, which means reads might not immediately reflect the most recent writes. This behavior explains why the application is not always receiving the latest data.

Because latency and performance are acceptable, the problem is not throughput or response time—it is data consistency. Therefore, the correct design change is to enable strongly consistent reads, which guarantee that reads always return the most up-to-date data after a successful write.

Option C is correct because strongly consistent reads ensure that the application sees the latest committed value for an item. This is critical for use cases such as inventory systems, user profiles, or transactional workflows where stale data can cause incorrect behavior.

Option A is invalid because DynamoDB does not support traditional read replicas. Option B does not affect consistency; GSIs are used to query data by alternative keys. Option D would worsen the problem because eventual consistency already causes stale reads.

Therefore, C is the appropriate recommendation to ensure correctness of returned data without changing the table schema or introducing additional complexity.

Why Option D is Correct:

GPU Access: Only EC2 instances in the GPU family (e.g., P2, P3) can provide GPU resources.

ECS Orchestration: Simplifies container deployment and management.

Why Other Options Are Not Ideal:

Option A: Lambda does not support GPU-based runtimes.

Option B: AWS Fargate does not support GPU-based workloads.

Option C: ECR is a container registry, not an orchestration or execution service.

AWS References:

Amazon ECS with GPU Instances:AWS Documentation - ECS GPU Instances