A meter-space wait scorer already runs on AWS Lambda. Cold starts and memory max-out are the complaint, and there is no SageMaker host to chart. What should they use?
Select an answer to reveal the explanation.
Short Explanation
The wait scorer already runs on Lambda, and the complaint is cold starts and memory max-out. Use CloudWatch Lambda Insights and the function's own metrics. SageMaker ModelLatency is a host chart that function does not publish.
Full Explanation
Lambda-hosted inference is observed with CloudWatch Lambda Insights and the function’s own metrics (duration, memory, cold starts). SageMaker ModelLatency is a host metric and is not published by that function. Rekognition is not Insights, and Insights does not require moving the workload onto a SageMaker multi-model endpoint.