A ferry-gate scoring endpoint is over its p99 budget. ModelLatency is about 40 ms, and OverheadLatency dominates. Staff want a larger instance. What should they do first?
Select an answer to reveal the explanation.
Short Explanation
ModelLatency is about 40 ms, and OverheadLatency dominates the p99 miss. Treat queueing or an extra container hop as the bottleneck. A larger instance will not fix that split.
Full Explanation
SageMaker splits ModelLatency from OverheadLatency. When the model itself is fast and overhead dominates, the fix is queueing, an extra container hop, or similar path cost, not a larger instance. Polly is not that split, and overhead is not a model-quality metric.