Which approach will meet these requirements with the LEAST operational overhead?
Configure an S3 event notification that invokes an AWS Lambda function when new documents are created. Configure the Lambda function to create three SageMaker batch transform jobs, one batch transform job for each model for each document.
Deploy all the models to a single SageMaker endpoint. Treat each model as a production variant. Configure an S3 event notification that invokes an AWS Lambda function when new documents are created. Configure the Lambda function to call each production variant and return the results of each model.
Deploy each model to its own SageMaker endpoint Configure an S3 event notification that invokes an AWS Lambda function when new documents are created. Configure the Lambda function to call each endpoint and return the results of each model.
Deploy each model to its own SageMaker endpoint. Create three AWS Lambda functions. Configure each Lambda function to call a different endpoint and return the results. Configure three S3 event notifications to invoke the Lambda functions when new documents are created.
Explanations:
Using three separate SageMaker batch transform jobs increases operational overhead and is less efficient for real-time predictions as batch jobs are typically not designed for low-latency use cases.
Deploying all models to a single SageMaker endpoint as production variants allows a streamlined, low-latency solution. A single Lambda function can call each variant, making it more efficient and reducing complexity.
Deploying each model to its own endpoint adds operational complexity and may increase latency. Calling multiple endpoints separately via Lambda results in higher overhead and is less efficient.
Creating three separate Lambda functions and S3 event notifications increases operational overhead and complexity. This approach is unnecessary and less efficient compared to a single Lambda function calling multiple production variants.