An EC2 fleet that runs nightly feature builds for a landfill model is about twice the vCPU it uses. Inference Recommender does not see those boxes. What should they use?
Select an answer to reveal the explanation.
Short Explanation
A nightly feature-build EC2 fleet is about twice the vCPU it uses. Inference Recommender does not see those boxes. Use AWS Compute Optimizer. Low utilization is a signal to shrink, not to add instances.
Full Explanation
AWS Compute Optimizer rightsizes EC2 and other supported ML-adjacent hosts. Inference Recommender is for SageMaker inference endpoints and does not see a nightly feature-build fleet. Polly is not that report, and low utilization is a signal to shrink, not to add instances.