You train a model and register it in your Azure Machine Learning workspace. You are ready to deploy the model as a real-time web service. You deploy the model to an Azure Kubernetes Service (AKS)…

Question

Options

• AModify the AKS service deployment configuration to enable application insights and re-deploy to
• BCreate an Azure Container Instances (ACI) web service deployment configuration and deploy the
• CAdd a breakpoint to the first line of the entry script and redeploy the service to AKS.
• DCreate a local web service deployment configuration and deploy the model to a local Docker
• ERegister a new version of the model and update the entry script to load the new version of the
• FCreate an Environment object based on the dependencies you specify in the environment yaml
• GCreate an inference configuration (InferenceConfig object) based on the environment and the

• A
  3% (2)
• B
  73% (48)
• C
  2% (1)
• D
  15% (10)
• E
  2% (1)
• G
  6% (4)

Explanation

Deploying to Azure Container Instances (ACI) is the recommended approach for debugging entry script errors iteratively. ACI is a lightweight, serverless container service that is much faster to deploy and update than AKS - a full ACI deployment takes minutes versus the overhead of AKS. You can rapidly fix the entry script, redeploy to ACI, and test the fix without the complexity of managing a Kubernetes cluster. ACI is specifically designed as a development/test deployment target in Azure ML before promoting to AKS production. Option A (enabling Application Insights) adds logging but doesn't simplify iteration. Option C (adding a breakpoint) doesn't work in remote deployments. Option D (local Docker) is viable for local debugging but requires Docker installed locally and doesn't reflect the AKS environment. Options E, F, and G address environment or model versioning issues rather than entry script debugging.

No community discussion yet for this question.