Answer-first summary for fast verification
Answer: Serverless compute
The question requires a compute target for Automated ML training that can automatically scale to a multi-node cluster to handle larger datasets. Serverless compute (Option D) is the optimal choice because it abstracts infrastructure management and automatically scales compute resources up or down based on job demand, specifically supporting multi-node clusters for large-scale Automated ML training. This aligns with Azure Machine Learning best practices for Automated ML, where serverless compute handles dynamic scaling without manual intervention. Other options are less suitable: Compute instance (A) is a single-node development environment without auto-scaling; Endpoints (B) are for deploying models, not training; Kubernetes cluster (C) requires manual configuration and management, lacking the automatic scaling capability needed here.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
No comments yet.
You are designing a training job in an Azure Machine Learning workspace using Automated ML. The compute resource must be able to scale up to a multi-node cluster automatically to handle larger datasets.
Which Azure Machine Learning compute target should you use?
A
Compute instance
B
Endpoints
C
Kubernetes cluster
D
Serverless compute