feat(report): record run_config (seeds + load/warmup) in result_summary.json

[nvzhihanj]

Summary

Record the run's load config + RNG seeds in the benchmark report so a run is self-describing and validatable from its own result_summary.json (and report.txt).

A single run_config field on Report carries:

• load_pattern (type / target_qps / concurrency / use_legacy_loadgen_qps_metrics — needed to interpret qps/tps),
• warmup (incl. warmup_random_seed),
• scheduler_random_seed + dataloader_random_seed.

load_pattern / warmup come from settings.model_dump(include={"load_pattern", "warmup"}); the seeds are read off settings.runtime. endpoint_config (api_key / endpoint URLs) is a sibling of settings and never included, so no secrets leak. The full, re-runnable config still lives in config.yaml alongside.

Replaces the standalone seeds field (#353): the RNG seeds now live under run_config.

Scope note: the resolved / effective runtime settings (the actual issued sample count and sample ordering, which can differ per audit phase — e.g. MLPerf TEST04's fixed-sample output-caching phase) are intentionally left to a follow-up PR. Getting those right means sourcing from the live rt_settings per phase, which is a separate change. This PR only records the static, run-identifying config above.

Example result_summary.json

From a real DGX-B300x8 DeepSeek-R1 server (poisson) run — qps/tps stay top-level, run config under run_config (metric dicts elided as { ... }):

{
  "version": "0.1.0",
  "git_sha": "...",
  "n_samples_completed": 40000,
  "duration_ns": 3328465383747,
  "state": "complete",
  "complete": true,
  "ttft": { ... }, "tpot": { ... }, "latency": { ... }, "output_sequence_lengths": { ... },
  "legacy_loadgen_window_duration_ns": 2753346633148,
  "qps": 14.53,
  "tps": 54489.98,
  "run_config":
  {
    "load_pattern": {
      "type": "poisson",
      "target_qps": 14.75,
      "target_concurrency": null,
      "use_legacy_loadgen_qps_metrics": true
    },
    "warmup": {
      "enabled": false,
      "n_requests": null,
      "salt": true,
      "drain": false,
      "warmup_random_seed": 42
    },
    "scheduler_random_seed": 42,
    "dataloader_random_seed": 42
  }
}

Same, rendered in report.txt:

Run config:
  load_pattern: type=poisson, target_qps=14.75, target_concurrency=None, use_legacy_loadgen_qps_metrics=True
  warmup: enabled=False, n_requests=None, salt=True, drain=False, warmup_random_seed=42
  scheduler_random_seed: 42
  dataloader_random_seed: 42

Tests

• run_config keyword-only passthrough; serialization + report.txt rendering; absent -> null.

Rebased on latest main (keeps PR #372's use_legacy_loadgen_qps_metrics; git_sha handled by #398).

🤖 Generated with Claude Code

[github-actions]

MLCommons CLA bot All contributors have signed the MLCommons CLA ✍️ ✅

[gemini-code-assist]

Code Review

This pull request introduces a mechanism to resolve the package's git commit SHA using git archive's export-subst feature via a new _git_archival.txt file, falling back to running git rev-parse anchored to the package directory. It also reorders the serialized JSON output of metrics reports to surface key throughput and run identity fields first. Feedback is provided regarding a potential issue where git rev-parse could traverse up to a parent repository if the package is installed in a virtual environment within an unrelated git repository, along with a suggestion to check for the existence of the .git directory first.

Important

The consumer version of Gemini Code Assist on GitHub is being sunset. Starting June 18, 2026, new organization installations will be blocked, and all code review activity will officially cease on July 17, 2026.
For more details on the timeline and next steps, please review the Help Documentation.