NVIDIA AI Infrastructure and Operations Operations and Monitoring NCA-AIIO
Quiz 8 Question 19 of 20

An operations team needs to deploy a containerized, multi-node deep learning inference application onto a shared Kubernetes cluster. The cluster contains a mix of GPU-enabled worker nodes and CPU-only general-purpose worker nodes. Which strategy best ensures the resource-intensive AI containers are scheduled onto the correct hardware while maintaining high availability and efficient resource scheduling?

Select an answer to reveal the explanation.