
Ultimate access to all questions.
Your company is planning to deploy a machine learning model as an Azure AI service. The model requires periodic retraining to maintain its accuracy. What strategies should you implement to ensure seamless retraining and deployment of the updated model without disrupting the service?
A
Disable the AI service during the retraining process and redeploy the updated model once the retraining is complete.
B
Implement a blue-green deployment strategy, where you maintain two identical environments, one for the current model and one for the updated model. Switch traffic between the environments during retraining.
C
Redeploy the updated model in the same environment as the current model, allowing them to run concurrently and gradually shift traffic to the updated model.
D
Ignore the need for retraining and continue using the initial model without any updates.