MLS-C01 Question 50
Select 4You are a Machine Learning Engineer working on a model training pipeline that processes large datasets in Amazon SageMaker. You want to schedule the training jobs to run automatically every day at midnight without manual intervention. Which AWS services or features can you use to schedule the SageMaker training jobs?
- A
Amazon EventBridge
- B
AWS Step Functions
- C
Amazon SageMaker Model Monitor
- D
AWS Lambda with a scheduled Amazon CloudWatch Events rule
- E
Amazon SageMaker Pipelines
Show answer and explanation
Correct answers: A, B, D, E
Explanation
To schedule jobs in Amazon SageMaker, you can use services designed for event-driven workflows or orchestration. Amazon EventBridge and AWS Step Functions provide mechanisms to automate and schedule training jobs. AWS Lambda can also be paired with CloudWatch Events (now part of EventBridge) to achieve the same goal. Additionally, SageMaker Pipelines is purpose-built for managing machine learning workflows, including scheduling tasks within a pipeline. SageMaker Model Monitor, on the other hand, is focused on monitoring and does not support job scheduling.
- A. Correct.
Amazon EventBridge is a serverless event bus service that can be used to schedule events, including invoking SageMaker training jobs on a regular basis.
- B. Correct.
AWS Step Functions can be used to orchestrate workflows, including scheduling and triggering SageMaker training jobs as part of a larger workflow.
- C. Incorrect.
Amazon SageMaker Model Monitor is used to monitor model performance and data drift, but it does not include job scheduling capabilities.
- D. Correct.
AWS Lambda, combined with a CloudWatch Events rule, can be used to trigger SageMaker training jobs based on a schedule.
- E. Correct.
Amazon SageMaker Pipelines allows you to build and orchestrate repeatable machine learning workflows, including scheduling training jobs.