You use an Azure Machine Learning workspace. You have a trained model that must be deployed as a web service. Users must authenticate by using Azure Active Directory. What should you do?

Question

Options

• ADeploy the model to Azure Kubernetes Service (AKS). During deployment, set the
• BDeploy the model to Azure Container Instances. During deployment, set the auth_enabled
• CDeploy the model to Azure Container Instances. During deployment, set the
• DDeploy the model to Azure Kubernetes Service (AKS). During deployment, set the

• A
  71% (17)
• B
  4% (1)
• C
  17% (4)
• D
  8% (2)

Explanation

The correct answer is A: Deploy to Azure Kubernetes Service (AKS) with token-based authentication enabled. Azure Active Directory authentication for ML web services is only supported on AKS deployments, not Azure Container Instances (ACI). AKS supports token-based auth (Azure AD tokens) by setting token_auth_enabled=True in the AksWebservice.deploy_configuration(). ACI only supports key-based authentication. Options B and C are incorrect because ACI does not support Azure AD authentication. The distinction is critical: ACI uses simple API keys, while AKS can integrate with Azure AD for enterprise-level token-based access control.

No community discussion yet for this question.