MLA-C01 Question 238
Select 3You are responsible for deploying a machine learning model that needs to handle real-time predictions with low latency. To manage the model's lifecycle, including versioning and monitoring, you choose to use AWS services. Which combination of services will best support your requirements?
- A
Amazon SageMaker Endpoint for real-time inference
- B
AWS Lambda for hosting the model and providing real-time predictions
- C
Amazon SageMaker Model Registry for versioning
- D
Amazon SageMaker Pipelines for workflow orchestration
- E
Amazon S3 for storing real-time inference requests and responses
Show answer and explanation
Correct answers: A, C, D
Explanation
To meet the need for real-time predictions with low latency, SageMaker Endpoints are the optimal choice for hosting the model. SageMaker Model Registry simplifies version management, while SageMaker Pipelines supports orchestration of deployment and monitoring workflows. This combination of services aligns with the requirements for deploying and managing the ML model effectively.
- A. Correct.
Amazon SageMaker Endpoint is designed for real-time inference, providing the low latency required in this scenario.
- B. Incorrect.
While AWS Lambda can host models, it is not optimized for hosting machine learning models with low latency. SageMaker Endpoints are a better fit.
- C. Correct.
Amazon SageMaker Model Registry is used for managing model versions, which is a critical requirement in this scenario.
- D. Correct.
Amazon SageMaker Pipelines helps orchestrate the ML workflow, including deployment and monitoring, making it suitable for the scenario.
- E. Incorrect.
Amazon S3 is a storage service but is not optimized for storing real-time inference requests and responses. It is not relevant for this specific requirement.