Question 31
Domain 4: ML Solution Monitoring, Maintenance, and SecurityAn ML engineer needs to deploy ML models to get inferences from large datasets in an asynchronous manner. The ML engineer also needs to implement scheduled monitoring of the data quality of the models. The ML engineer must receive alerts when changes in data quality occur. Which solution will meet these requirements?
Correct answer: D
Explanation
Amazon SageMaker batch transform is designed for asynchronous inference on large datasets, matching the need to "get inferences from large datasets in an asynchronous manner." SageMaker Model Monitor can "monitor the data quality" on a schedule and trigger alerts when drift or quality changes are detected, satisfying the monitoring and notification requirements.
Why each option is right or wrong
A. Deploy the models by using scheduled AWS Glue jobs. Use Amazon CloudWatch alarms to monitor the data quality and to send alerts.
AWS Glue is primarily an ETL/data integration service, not the standard managed choice for ML batch inference.
B. Deploy the models by using scheduled AWS Batch jobs. Use AWS CloudTrail to monitor the data quality and to send alerts.
CloudTrail records API activity for auditing; it does not evaluate model data quality drift.
C. Deploy the models by using Amazon Elastic Container Service (Amazon ECS) on AWS Fargate. Use Amazon EventBridge to monitor the data quality and to send alerts.
ECS on Fargate can run containers, but EventBridge is for event routing and scheduling, not data quality analysis.
D. Deploy the models by using Amazon SageMaker batch transform. Use SageMaker Model Monitor to monitor the data quality and to send alerts.
Amazon SageMaker batch transform is the SageMaker inference mode intended for offline, asynchronous scoring of large datasets, so it fits the requirement to process data without a real-time endpoint. SageMaker Model Monitor can be configured with a monitoring schedule and baseline constraints to evaluate data quality over time; when violations are detected, it publishes findings to Amazon CloudWatch, which can trigger alerts through alarms or notifications.