Answer-first summary for fast verification
Answer: Azure Machine Learning compute cluster
The question specifies a batch inference pipeline that needs to be scheduled to run nightly, processing multiple data files. Azure Machine Learning compute cluster (option B) is the optimal choice because it is designed for scalable, parallel batch workloads and supports job scheduling. It can automatically scale up and down based on workload, making it cost-effective for scheduled batch jobs. In contrast, Azure Machine Learning compute instance (option A) is intended for development and interactive work, not production batch jobs. Azure Kubernetes Service (AKS, option C) and Azure Container Instance (ACI, option D) are primarily for real-time (online) inference, not batch processing. The community discussion, with the highest upvoted comments (e.g., 10 upvotes for BTAB's comment) and references to official documentation (e.g., the example notebook link), strongly supports B as the correct answer, emphasizing the distinction between batch (compute cluster) and real-time (AKS/ACI) inference targets.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
You have a trained and registered machine learning model. You create a batch inference pipeline that uses this model to generate predictions from multiple data files. You plan to publish this batch inference pipeline as a service that can be scheduled to run nightly.
Which compute target should you select for the inference service?
A
Azure Machine Learning compute instance
B
Azure Machine Learning compute cluster
C
Azure Kubernetes Service (AKS)-based inference cluster
D
Azure Container Instance (ACI) compute target
No comments yet.