AI-900 Practice Question: Describe fundamental principles of machine learning on Azure

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  A scalable cluster for running distributed training jobs across many nodes

  Why it's wrong here

  This option describes an Azure Machine Learning compute cluster, not a compute instance. A compute instance is a single-node development VM with pre-installed tools like Jupyter and the Azure ML SDK, whereas a compute cluster is specifically designed to scale across many CPU/GPU nodes for distributed training jobs. The question asks about the interactive workstation for individual data scientists, not the parallel training infrastructure that a cluster provides.

• ✓

  A managed cloud workstation for interactive ML development with pre-installed tools

  Why this is correct

  A compute instance is a managed, single-node cloud workstation pre-configured with Azure Machine Learning Studio, Python environments, notebooks, and common development tools such as VS Code. Data scientists use it interactively to write code, explore data, prototype models, and run experiments without manually provisioning or configuring a virtual machine. This is exactly the definition of an Azure Machine Learning compute instance, so this is the correct answer.

• ✗

  A virtual machine that automatically scales to run batch predictions

  Why it's wrong here

  This describes batch inference infrastructure, not a compute instance. A compute instance is a fixed-size, single VM allocated for interactive development and does not automatically scale to process batch scoring workloads. Batch predictions are typically executed via batch endpoints or compute clusters that apply autoscaling rules, so this option conflates the development workstation (compute instance) with the separate infrastructure used for asynchronous scoring jobs.

• ✗

  A serverless execution environment for ML inference requests

  Why it's wrong here

  This option refers to Azure Machine Learning managed online endpoints, which can provision serverless compute, scale to zero, and bill per request for real-time inference. A compute instance is a persistent, interactive VM that you explicitly start and stop, and it is not designed to serve HTTP inference requests as a serverless execution environment. Therefore, this answer describes managed endpoints rather than the compute instance that the question is asking about.

Go deeper

Related to this question