AI-300 Certification Practice Question #3

Question

A team manages an Azure Machine Learning workspace where they deploy models to online endpoints.
The team needs to introduce a new version of a model to production without disrupting existing users.
The team must validate the new version before full rollout.
You need to reduce risk during deployment.
What should you do?

Accepted Answer

Introducing a new model version *without disrupting existing users* and *validating before full rollout* is exactly the blue/green safe-rollout pattern. Azure ML online endpoints let you add the new deployment and allocate a small traffic percentage to it, then increase gradually once validated. Replacing the endpoint or routing 100% of traffic to the new version (C, D) removes the safety net, and a batch endpoint (A) doesn't serve the real-time traffic in question.

More AI-300 practice questions