AI-900 Question 137
Single answerYou are a data scientist tasked with deploying a trained machine learning model using Azure Machine Learning. Your organization requires the deployment to handle real-time predictions and provide scalability as demand increases. Which deployment option should you choose in Azure Machine Learning to meet these requirements?
- A
Azure Container Instances (ACI)
- B
Azure Kubernetes Service (AKS)
- C
Azure Batch Inference
- D
Azure Functions
Show answer and explanation
Correct answer: B
Explanation
Azure Kubernetes Service (AKS) is the most appropriate choice for deploying machine learning models that require real-time inference and scalability. It supports containerized deployments, auto-scaling, and integration with Azure Machine Learning, making it ideal for high-demand production scenarios.
- A. Incorrect.
Azure Container Instances (ACI) is suitable for low-scale, development, or testing purposes but is not ideal for handling scalable, real-time predictions.
- B. Correct.
Azure Kubernetes Service (AKS) is designed for real-time inference at scale. It provides scalability and is the recommended option for handling high-demand scenarios.
- C. Incorrect.
Azure Batch Inference is used for large-scale, asynchronous batch processing of data, not real-time predictions.
- D. Incorrect.
Azure Functions is a serverless compute service that can be used for lightweight tasks, but it is not specifically designed for deploying machine learning models or handling scalable real-time predictions.