Question

A logistics company runs a fraud-detection model that must score every incoming transaction within 100 milliseconds, 24/7, with steady and predictable traffic of roughly 500 requests per second. The ML engineering team needs to choose the SageMaker deployment option that meets the latency requirement while keeping the infrastructure fully managed. Which deployment approach best fits these requirements?

Accepted Answer

Deploy the model to a SageMaker real-time endpoint backed by a fixed fleet of instances behind an auto scaling policy.

Answer

Use SageMaker Batch Transform on a scheduled basis to score accumulated transactions.

Answer

Deploy the model as a SageMaker Serverless Inference endpoint to avoid managing instances.

Answer

Use SageMaker Asynchronous Inference to queue and process transactions as they arrive.