An ML engineer needs to encrypt all data in transit when an ML training job runs. The ML engineer must ensure that encryption in transit is applied to processes that Amazon SageMaker uses during the t

Question

Options

• AEncrypt communication between nodes for batch processing.
• BEncrypt communication between nodes in a training cluster.
• CSpecify an AWS Key Management Service (AWS KMS) key during creation of the training job
• DSpecify an AWS Key Management Service (AWS KMS) key during creation of the SageMaker

• A
  3% (1)
• B
  93% (37)
• C
  5% (2)

Explanation

Option B is correct because SageMaker distributed training jobs communicate across multiple compute nodes in a cluster, and enabling inter-node (inter-container) traffic encryption directly addresses encryption in transit - exactly what the question requires. SageMaker exposes this as a specific EnableInterContainerTrafficEncryption flag when creating a training job.

Option A is wrong because "batch processing" refers to SageMaker Batch Transform (used for inference), not training jobs - it's the wrong SageMaker feature entirely.

Options C and D are wrong because specifying an AWS KMS key encrypts data at rest (storage volumes, model artifacts, S3 output), not data in transit between nodes. KMS keys are the answer to an "at-rest" encryption question, not a "in-transit" question.

Memory tip: Use the phrase "In Transit = In the Cluster." When data moves between training nodes, it's traveling - that's in-transit encryption. When data sits on a disk or in S3, that's at-rest encryption (KMS). The moment you see "training job + in transit," think inter-node/inter-container traffic encryption.

No community discussion yet for this question.