Skip to content

RoCE Configuration and Operations on Spectrum-X · Part 7 of 9

Act 7 ”” CloudAI Benchmark for Performance Baselining

The NVIDIA CloudAI benchmark is the industry-standard acceptance test for Spectrum-X deployments.

What CloudAI measures:

TestMetricPass threshold (per port)
All-reduce bandwidthAggregate BW across all GPUs≥ 380 GB/s (8-GPU DGX)
All-to-all bandwidthIncast stress test≥ 360 GB/s
Point-to-pointSingle-flow line rate≥ 46.5 GB/s
Latency distributionp50, p99, p999p99 ≤ 2µs, p999 ≤ 10µs
FairnessCV (coefficient of variation)CV < 5% across flows

Running the benchmark:

dgx-01:~$ cloudai --mode run \
  --system-config configs/spectrum-x-8node.toml \
  --test-scenario configs/tests/nccl-all-reduce.toml

[cloudai] All-reduce BW: 382.1 GB/s avg (CV=0.6%) ”” PASS
[cloudai] Baseline stored: /results/cloudai/baseline-2024-01-15.json

Comparison against baseline:

dgx-01:~$ cloudai --mode compare \
  --baseline /results/cloudai/baseline-2024-01-15.json \
  --current /results/cloudai/run-2024-01-22.json

[cloudai] All-reduce BW: 382.1 → 383.4 GB/s (+0.3%) ”” OK
[cloudai] p99 latency:   1.21  → 1.18 µs   (-2.5%) ”” OK
[cloudai] Overall: NO REGRESSION DETECTED