Free AWS SAA-C03 Actual Exam Questions – Solutions Architect Questions - Question 13 Discussion
during the company's weekly sales events. The application's backend uses an Amazon API Gateway
HTTP API and AWS Lambda functions to process user requests. During peak sales periods, users
report encountering TooManyRequestsException errors from the Lambda functions. The errors result
in a degraded user experience. A solutions architect needs to design a scalable and resilient solution
that minimizes the errors and ensures that the application's overall functionality remains unaffected.
Option B could work too since Step Functions help manage workflows and can throttle requests by controlling the execution flow, which might reduce Lambda overload without relying solely on concurrency limits.
D, because reserved concurrency directly controls max Lambda instances, avoiding throttling.
Maybe D is better because reserved concurrency directly limits how many Lambda instances run at once, preventing the function from getting swamped. The SQS queue smooths out the burst of requests by holding them until the Lambda can process them. Provisioned concurrency in A helps with cold starts but doesn’t control throttling as well, so you could still get TooManyRequestsException errors if the traffic is really high. So D seems more about controlling the load, not just keeping Lambdas warm.
D. The key here is that reserved concurrency can help throttle the Lambda invocations to a set limit, preventing the function from being overwhelmed and throwing TooManyRequestsException errors. Pairing it with SQS lets you queue requests and process them smoothly without losing traffic. Provisioned concurrency (A) mainly addresses cold starts but doesn't directly limit concurrency, so it might not fully solve the throttling issue during traffic spikes. Also, SNS (C) isn’t ideal here since it pushes all messages simultaneously, which could worsen throttling. Step Functions (B) adds orchestrati
Makes sense to go with D since reserved concurrency controls Lambda limits better under load. D
Maybe D works better here since reserved concurrency caps execution to prevent Lambda from being overwhelmed, and SQS smooths spikes by queuing requests. Provisioned concurrency (A) helps with cold starts but might not control throttling as well.
Option A makes sense because using provisioned concurrency keeps the Lambda ready to handle bursts without cold starts, which helps during those heavy Monday spikes. Plus, SQS helps buffer the load so requests don’t overwhelm Lambda all at once.
Option D is close but reserved concurrency caps the number of concurrent executions, which might still cause throttling if the queue grows too fast. Provisioned concurrency in A ensures capacity upfront, which seems better for minimizing TooManyRequestsException errors during peak loads.
A/D? Need clarity on concurrency limits and how they differ between provisioned and reserved.