AI-900 Question 140
Single answerYou are developing a machine learning model using Azure Machine Learning and need to deploy it to a production environment. The deployment must allow you to monitor the model's performance, manage scaling, and perform version control. Which deployment option in Azure Machine Learning would best meet these requirements?
- A
Deploy the model as a web service in an Azure Kubernetes Service (AKS) cluster
- B
Deploy the model to an Azure Blob Storage container
- C
Deploy the model as a batch inference pipeline
- D
Deploy the model to a local environment for testing
Show answer and explanation
Correct answer: A
Explanation
Azure Kubernetes Service (AKS) is the recommended deployment option for production scenarios in Azure Machine Learning. It allows you to expose your model as a web service, scale based on demand, monitor its performance, and manage different versions of the model. Other options, like Blob Storage or local environments, do not provide the necessary production-level features.
- A. Correct.
Deploying the model as a web service in an Azure Kubernetes Service (AKS) cluster provides features like scaling, monitoring, and version control, making it suitable for production environments.
- B. Incorrect.
Azure Blob Storage is used for storing data and files, but it does not provide the capabilities needed for deploying and managing machine learning models.
- C. Incorrect.
A batch inference pipeline is suitable for processing large volumes of data in batches, but it is not ideal for real-time production deployments requiring scaling and monitoring.
- D. Incorrect.
Deploying the model to a local environment is useful for testing but does not support production-level features like scaling and monitoring.