You use an Azure Machine Learning workspace. You must monitor cost at the endpoint and deployment level. You have a trained model that must be deployed as an online endpoint. Users must authenticate…

Question

Options

• ADeploy the model to Azure Kubernetes Service (AKS). During deployment, set the
• BDeploy the model to Azure Kubernetes Service (AKS). During deployment, set the auth_mode
• CDeploy the model to a managed online endpoint. During deployment, set the auth_mode
• DDeploy the model to a managed online endpoint. During deployment, set the token_auth_mode

• A
  15% (4)
• B
  7% (2)
• C
  74% (20)
• D
  4% (1)

Explanation

Managed online endpoints in Azure Machine Learning natively support cost monitoring broken down by endpoint and individual deployment. They support Microsoft Entra ID (formerly Azure AD) authentication by setting auth_mode to 'aad_token'. AKS deployments (options A and B) do not provide the built-in endpoint/deployment-level cost monitoring that managed online endpoints offer. Option D is incorrect because 'token_auth_mode' is not a valid parameter name for configuring Entra ID authentication on managed online endpoints.

No community discussion yet for this question.