Straggler Detection

The Straggler Detection package’s purpose is to detect slower ranks participating in a PyTorch distributed workload. The nvidia-resiliency-ext package also includes the PTL callback StragglerDetectionCallback that simplifies integration with PyTorch Lightning-based workloads.

Straggler Detection is included in the nvidia_resiliency_ext.straggler package. StragglerDetectionCallback is included in the nvidia_resiliency_ext.ptl_resiliency package.

Contents:

• Usage guide
  • Straggler Detection Package Design Overview
  • Straggler Detection Usage Overview
  • Straggler Detection Integration Guide
• API documentation
  • Straggler
  • Reporting
  • Statistics
  • Callback
• Examples
  • Basic usage example