QuestionQ87

VerifiedDeployment and Orchestration of ML Workflows

A company uses Amazon SageMaker to develop ML models. The company keeps sensitive training data in an Amazon S3 bucket. Model training must be network-isolated from the internet.

Which solution meets this requirement?

Explanation

SageMaker training jobs can run in private subnets without internet access. An Amazon S3 gateway VPC endpoint adds private routing to S3, allowing the jobs to access training data without using a NAT gateway or traversing the public internet. AWS documents this configuration for training containers that need S3 access while the VPC has no internet access.

Learn more

Community Discussion

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