2.4 Leveraging Gemini Enterprise Agent Platform for End-To-End ML Workflows PROFESSIONAL-CLOUD-ARCHITECT Practice Questions (Page 6)

Questions 26–29

1. 26foundation · easy

   How does Agent Platform Pipelines orchestrate the execution of components in a pipeline?

   • ABy executing components in a predefined sequence, respecting dependencies and passing data between them.
   • BBy running all components simultaneously and merging their outputs at the end.
   • CBy requiring a human to manually trigger each component in order.
   • DBy using a single large container that contains all the code for every step.
   Reveal Solution Discussion Select an answer first
2. 27foundation · easy

   What is the primary purpose of Agent Platform Pipelines in an ML lifecycle?

   • ATo automate and orchestrate the steps of an ML workflow, from data preparation through model deployment.
   • BTo provide a managed environment for running Jupyter notebooks interactively.
   • CTo monitor the performance of deployed models in production.
   • DTo store and version trained ML models in a central repository.
   Reveal Solution Discussion Select an answer first
4. 28foundation · easy

   What is AI Hypercomputer in the context of Google Cloud?

   • AA supercomputer architecture designed specifically for AI/ML workloads, integrating hardware, software, and networking.
   • BA data storage service optimized for AI training datasets.
   • CA managed service for deploying ML models to mobile devices.
   • DA single powerful VM that can run any ML model.
   Reveal Solution Discussion Select an answer first
5. 29foundation · easy

   How can Cloud Run functions be used alongside Agent Platform in ML/AI workloads?

   • AAs a replacement for Agent Platform Pipelines.
   • BAs event-driven triggers that start pipeline runs or process data before it enters the pipeline.
   • CAs the primary compute for training large ML models.
   • DAs a storage service for ML training data.
   Reveal Solution Discussion Select an answer first