4.12 HPL burn-in

Key points

1. HPL (High-Performance Linpack) solves a large dense linear system. It loads the GPUs hard and also uses the network. NVIDIA's SuperPOD admin guide lists NCCL for the fabric and HPL for math-intensive applications with network communication.

   What NVIDIA says (2)

   “Math intensive applications with network communications”

   — DGX SuperPOD Administration Guide: System Health Checks and Debugging

   “The NVIDIA HPL benchmark expects one GPU per MPI process.”

   — NVIDIA HPC Benchmarks: NVIDIA HPL Benchmark

2. HPL.dat is HPL's input file. It sets the problem size and process grid. NVIDIA ships samples of both input files and Slurm batch-job scripts with the package.

   What NVIDIA says (2)

   “Samples of Slurm batch-job scripts in sample-slurm directory”

   — NVIDIA HPC Benchmarks: NVIDIA HPL Benchmark

   “Samples of input files in sample-dat directory”

   — NVIDIA HPC Benchmarks: NVIDIA HPL Benchmark

Key terms

• High-Performance Linpack: A math-heavy benchmark used to load and compare systems; NVIDIA HPL runs one GPU per MPI process.
• Burn-in: Running a heavy workload for a long time to expose weak parts before production.

Sample question

Why is HPL used for burn-in alongside NCCL?