A company is planning to create several ML prediction models. The training data is stored in Amazon S3. The entire dataset is more than 5 ТВ in size and consists of CSV, JSON, Apache Parquet, and simp

Question

Options

• AProcess data at each step by using Amazon SageMaker Data Wrangler. Automate the process by
• BUse Amazon SageMaker notebooks for each data processing step. Automate the process by
• CProcess data at each step by using AWS Lambda functions. Automate the process by using AWS
• DUse Amazon SageMaker Pipelines to create a pipeline of data processing steps. Automate the

• A
  2% (1)
• B
  6% (3)
• C
  13% (6)
• D
  79% (37)

Explanation

SageMaker Pipelines (D) is purpose-built for orchestrating multi-step ML workflows end-to-end. It natively supports the full range of data formats stored in S3 (CSV, JSON, Parquet, text), handles long-running processing steps without timeout constraints, integrates with SageMaker Processing for NLP transformations, and provides built-in automation and scheduling - exactly matching all stated requirements.

Why the distractors fail:

• A (Data Wrangler): Designed for interactive, GUI-driven data preparation - not for automating complex, multi-hour, multi-step pipelines in production.
• B (SageMaker Notebooks): Notebooks are for interactive exploration and development, not for reliable automated orchestration of consecutive processing steps.
• C (Lambda): Lambda has a maximum execution timeout of 15 minutes, making it fundamentally unsuitable for processing steps that take hours to complete.

Memory tip: Think of SageMaker Pipelines as the "assembly line" for ML - when you see keywords like consecutive steps, hours to finish, automated, and multi-format data at scale, that's the pipeline pattern. Lambda's 15-minute hard limit is a common exam trap for long-running workloads.

No community discussion yet for this question.