MLS-C01 Question 357
Select 2Your team has trained a machine learning model to predict customer churn and now needs to deploy it into production for real-time inference. The team wants to ensure the deployment is highly available, can scale automatically with traffic, and allows for easy rollback in case of issues. Which AWS services and features should you use to meet these requirements?
- A
Deploy the model using Amazon SageMaker Endpoint with Auto Scaling enabled.
- B
Use Amazon Lambda to deploy the model and configure Provisioned Concurrency.
- C
Leverage Amazon EC2 Auto Scaling Groups for model deployment and scaling.
- D
Enable Amazon SageMaker Multi-Model Endpoints to support rollback to previous models.
- E
Use Amazon SageMaker Endpoint Configurations to create and manage multiple model versions.
Show answer and explanation
Correct answers: A, E
Explanation
To deploy and operationalize ML solutions effectively, Amazon SageMaker Endpoints with Auto Scaling provide a scalable and highly available environment for real-time inference. Combined with SageMaker Endpoint Configurations, you can manage multiple model versions and easily perform rollbacks, ensuring robust production deployments.
- A. Correct.
Correct: Amazon SageMaker Endpoints with Auto Scaling provide a managed and scalable environment for deploying machine learning models in production. It ensures high availability and can automatically adjust resources based on traffic.
- B. Incorrect.
Incorrect: While AWS Lambda can be used for deploying models for inference, it is not designed for handling large-scale, real-time inference workloads. Provisioned Concurrency can improve performance but does not inherently support rollback or version management for models.
- C. Incorrect.
Incorrect: EC2 Auto Scaling Groups can scale compute resources, but they require significant custom setup and do not natively support machine learning model deployment or rollback.
- D. Incorrect.
Incorrect: SageMaker Multi-Model Endpoints are designed for hosting multiple models on the same endpoint to optimize cost but are not specifically for rolling back to previous models.
- E. Correct.
Correct: SageMaker Endpoint Configurations allow you to manage multiple versions of a deployed model, making it easier to switch to a previous version for rollback in case of issues.