GPU Workloads in Containers Quiz

5 questions Pass: 70% +25 pts (50 with Pro)

Quiz covering Advanced Container Patterns

What you'll be asked

  1. Which component is primarily responsible for exposing host GPU resources to a containerized application?
  2. When deploying a GPU-enabled workload in Kubernetes, which field must be specified in the container resource limits to request a GPU?
  3. Why is it generally recommended to use a base image provided by the GPU vendor (e.g., nvidia/cuda) instead of a generic OS image for GPU workloads?
  4. What is the primary purpose of the 'NVIDIA_VISIBLE_DEVICES' environment variable in a container?
  5. In a scenario where multiple containers in the same pod require GPU access, how does the NVIDIA device plugin handle resource allocation?

Answers and explanations come after you submit.