A company needs to create a central catalog for all the company's ML models. The models are in AWS accounts where the company developed the models initially. The models are hosted in Amazon Elastic…

Question

Options

• AConfigure ECR cross-account replication for each existing ECR repository. Ensure that each
• BCreate a new AWS account with a new ECR repository as the central catalog. Configure ECR
• CUse the Amazon SageMaker Model Registry to create a model group for models hosted in
• DUse an AWS Glue Data Catalog to store the models. Run an AWS Glue crawler to migrate the

• A
  2% (1)
• B
  7% (4)
• C
  87% (47)
• D
  4% (2)

Explanation

Amazon SageMaker Model Registry is purpose-built for exactly this use case - cataloging, versioning, and managing ML models across an organization. It supports models stored in ECR and can serve as a centralized registry across multiple AWS accounts without requiring any physical migration of the underlying container images.

Why the distractors fail:

• A & B both involve ECR replication or migration, which duplicates image data across accounts rather than creating a logical catalog - they solve the wrong problem (data movement, not model governance).
• D AWS Glue Data Catalog is designed for structured data assets (tables, schemas, databases in a data lake), not ML model artifacts; it has no concept of model versions, approval workflows, or inference containers.

Memory tip: Match the service to the domain - Glue catalogs data, SageMaker catalogs models. Whenever an exam question involves "ML model management, versioning, or governance across accounts," SageMaker Model Registry is almost always the answer over generic AWS data catalog services.

No community discussion yet for this question.