MLA-C01 Question 352
Single answerYou are deploying a new version of a machine learning model to production using Amazon SageMaker. You want to minimize downtime and ensure that you can quickly roll back to the previous version in case of any issues. Which deployment strategy would be the most appropriate to achieve this?
- A
Blue/green deployment
- B
Canary deployment
- C
Linear deployment
- D
Rolling deployment
Show answer and explanation
Correct answer: A
Explanation
Blue/green deployment is ideal for scenarios where minimizing downtime and enabling quick rollback are critical. By maintaining two separate environments (blue and green), you can test the new model in the green environment without disrupting traffic to the blue environment. If the new model has issues, rollback is as simple as directing traffic back to the blue environment.
- A. Correct.
Blue/green deployment is designed to minimize downtime by deploying the new version in a separate environment (green) while keeping the current version (blue) running. It also allows for an easy rollback by switching traffic back to the blue environment if issues are detected.
- B. Incorrect.
Canary deployment introduces the new model to a small portion of the traffic initially, gradually increasing traffic over time. While it minimizes risk, it does not provide the immediate rollback capability that blue/green deployment offers.
- C. Incorrect.
Linear deployment involves gradually shifting traffic from the old version to the new version in equal increments over time. While this reduces risk, it is not as focused on rapid rollback as blue/green deployment.
- D. Incorrect.
Rolling deployment is typically used for updating applications running on multiple instances by updating them one batch at a time. This strategy is not commonly used for machine learning model deployments in SageMaker.