AWS Certified Machine Learning – SpecialtyMachine Learning Implementation and OperationsEasy

A financial institution is developing a new credit risk assessment model. They need a centralized, versioned, and low-latency storage solution for features that can be used consistently across both model training and real-time inference. The solution must support point-in-time correctness for historical training data and fast retrieval for online predictions. Which SageMaker service is designed to meet these requirements?

  1. AAmazon S3 for raw data storage.
  2. BSageMaker Feature Store.
  3. CSageMaker Model Registry for model versioning.
  4. DAmazon DynamoDB for feature storage.
Show answer & explanation

Correct answer: B. SageMaker Feature Store.

SageMaker Feature Store is specifically designed to address the challenges of feature management for ML, providing a centralized repository for features. It supports both online (low-latency retrieval for inference) and offline (point-in-time correctness for training) access, along with versioning and consistent feature definitions.

Why the other options are wrong

  • A. Amazon S3 is a general-purpose object storage and lacks native support for online/offline feature serving, versioning, and point-in-time correctness specifically for ML features.
  • C. SageMaker Model Registry is for managing and versioning *models*, not raw or engineered *features*.
  • D. Amazon DynamoDB can provide low-latency storage but lacks the built-in MLOps capabilities like offline store, point-in-time correctness, and feature versioning that SageMaker Feature Store offers.

SageMaker Feature Store

A purpose-built service for MLOps that provides a centralized repository for storing, discovering, and sharing ML features for both training and inference.

  • Includes an online store for low-latency inference and an offline store for training.
  • Ensures feature consistency between training and inference (prevents skew).
  • Supports point-in-time lookups for historical feature values.

Memory trick: Features, features, everywhere, a central place to show you care!

More Machine Learning Implementation and Operations questions