QuestionQ359

Data Security and Governance

A data engineer must share a dataset containing customer transaction data with a machine learning (ML) team in a separate AWS account. The dataset is stored in Amazon S3 and includes sensitive information that requires governance controls. The data engineer needs to ensure that the ML team can discover the dataset and request access to it. The solution must preserve appropriate security controls and track data lineage.

Which solution meets these requirements?

  • A Create cross-account IAM roles that grant the ML team direct access to the S3 bucket where the data is stored. Use AWS CloudTrail to track data access.
  • B Configure an Amazon SageMaker Unified Studio data catalog project that contains the dataset with appropriate metadata and project-based access controls.
  • C Set up cross-account S3 bucket replication to copy the dataset to the ML team's account. Use S3 server access logging to monitor usage.
  • D Create an AWS Lake Formation data catalog. Define tag-based access controls that allow the ML team to query the data directly from the team's account.
Explanation

Amazon SageMaker Unified Studio provides a governed business data catalog in which data assets can be enriched with metadata, published for discovery across projects, and accessed through subscription requests that can require owner approval. Approved subscriptions grant the required Lake Formation permissions, and catalog asset details include lineage, satisfying governed access, discoverability, request-based access, and lineage requirements.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!