MLA-C01 Question #45: Real Exam Question with Answer & Explanation

Question

A company has trained and deployed an ML model by using Amazon SageMaker. The company needs to implement a solution to record and monitor all the API call events for the SageMaker endpoint. The solution also must provide a notification when the number of API call events breaches a threshold. Which solution will meet these requirements?

Options

• AUse SageMaker Debugger to track the inferences and to report metrics. Create a custom rule to
• BUse SageMaker Debugger to track the inferences and to report metrics. Use the tensor_variance
• CLog all the endpoint invocation API events by using AWS CloudTrail. Use an Amazon
• DAdd the Invocations metric to an Amazon CloudWatch dashboard for monitoring. Set up a

Explanation

Option C is correct because AWS CloudTrail is specifically designed to record API call events across AWS services, including SageMaker endpoint invocations - it captures who called what, when, and from where. Pairing CloudTrail with Amazon CloudWatch (implied by the truncated answer) allows you to create metric filters and alarms that trigger notifications (via SNS) when invocation counts breach a threshold.

Why the distractors fail:

• A & B both involve SageMaker Debugger, which is a training-time tool for capturing tensors, gradients, and debugging model training - it has no role in monitoring endpoint API call events post-deployment.
• D uses the CloudWatch Invocations metric, which tracks inference request counts but does not record individual API call events (caller identity, request details, audit trail). This satisfies monitoring but misses the "record events" requirement, which needs CloudTrail.

Memory tip: Think "CloudTrail = audit log of API calls" - any time an exam question says "record API call events" or "audit who called what," CloudTrail is almost always the answer. CloudWatch handles metrics and alarms, CloudTrail handles who did what.

No community discussion yet for this question.