Q98Design Resilient Architectures
A company has an application with a REST-based interface that enables data to be received in near real time from a third-party vendor. After the data is received, the application processes it and stores it for additional analysis. The application runs on Amazon EC2 instances. The third-party vendor has encountered many **503 Service Unavailable** errors while sending data to the application. When data volume spikes, compute capacity reaches its maximum limit, and the application cannot process all requests. Which design should a solutions architect recommend to provide a more scalable solution?
← → navigate · a answer
Community votes
Discussion · 6
A 7
The key reasons are:
Kinesis Data Streams provides an auto-scaling stream that can handle large amounts of streaming data ingestion and throughput. This removes the bottlenecks around receiving the data.
AWS Lambda can process and store the data in a scalable serverless manner, avoiding EC2 capacity limits.
API Gateway adds API management capabilities but does not improve the underlying scalability of the EC2 application.
SNS is for event publishing/notifications, not large scale data ingestion. ECS still relies on EC2 capacity.
A 5
For near-real time data ingest and processing, Kinesis and Lambda are most scalable choice.
3
A.
Kinesis Data Streams = near realtime and scalable
AWS Lambda functions = scalable
A 3
more scalable solution? = serverless = Amazon Kinesis Data Streams and AWS Lambda functions
2
Only A is pure serverless which means scale. A for sure.
B 1
A change the whole architect, B adding a API gateway to make the app more scalable not perfect as in A it may not be possible within budget, deadline, Yes, AWS API Gateway significantly contributes to making an application more scalable by handling a large volume of concurrent API requests, automatically managing traffic distribution, it can cache response and allow asyn call