QuestionQ70

VerifiedDeployment and Orchestration of ML Workflows

A company routinely receives new training data from the vendor of an ML model. Every 3–4 days, the vendor places cleaned and prepared data in the company’s Amazon S3 bucket.

The company uses an Amazon SageMaker pipeline to retrain the model. An ML engineer must implement a solution that runs the pipeline whenever new data is uploaded to the S3 bucket.

Which solution meets these requirements with the LEAST operational effort?

Explanation

Amazon S3 can send object-created events to Amazon EventBridge, and Amazon SageMaker Pipelines is a supported EventBridge target. An EventBridge rule can match the S3 upload event and directly start a pipeline execution, eliminating custom polling code and the operational overhead of managing an Airflow environment.

Learn more

Community Discussion

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