Topology-Aware GPU Scheduling

Topology-Aware Scheduling

Learn how NVIDIA Run:ai uses topology-aware scheduling to optimize AI workload performance.

Note

This video was recorded using NVIDIA Run:ai version 2.24.18. The user interface, features, and workflows may differ in newer releases. For the latest information, refer to the current documentation.

What You'll Learn:

• Understand how topology-aware scheduling improves workload placement

• Optimize GPU placement for AI workloads

• Use NVLink connectivity to improve communication efficiency

• Account for node topology when scheduling workloads

• Improve performance for GPU-intensive AI workloads