Michael Chus 0d925299ff Use per-GPU temperature limits from nvidia-smi -q for headroom calculation ...

Parse "GPU Shutdown Temp" and "GPU Slowdown Temp" from nvidia-smi -q verbose
output in enrichGPUInfoWithMaxClocks. Store as ShutdownTempC/SlowdownTempC
on benchmarkGPUInfo and BenchmarkGPUResult. Fallback: 90°C shutdown / 80°C
slowdown when not available.

TempHeadroomC = ShutdownTempC - P95TempC (per-GPU, not hardcoded 100°C).
Warning threshold: p95 >= SlowdownTempC. Critical: headroom < 10°C.
Report table shows both limits alongside headroom and p95 temp.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

2026-04-16 06:45:15 +03:00

audit

Use per-GPU temperature limits from nvidia-smi -q for headroom calculation

2026-04-16 06:45:15 +03:00

bible @ 1d89a4918e

Update bible submodule

2026-04-08 07:14:31 +03:00

bible-local

Split bee-bench into perf and power workflows

2026-04-14 17:33:13 +03:00

internal

chore: commit pending repo changes

2026-03-31 22:17:36 +03:00

iso

Tune bee-gpu-burn single-precision benchmark phases

2026-04-16 00:05:47 +03:00

scripts

Drop legacy non-container builders

2026-03-16 00:23:55 +03:00

.env.example

fix: run-builder.sh uses BUILDER_USER from .env, not hardcoded

2026-03-08 18:48:33 +03:00

.gitignore

Disable precision fallback and pin cuBLAS 13.1

2026-04-14 10:17:44 +03:00

.gitmodules

Embed Reanimator Chart web viewer

2026-03-15 22:07:42 +03:00

PLAN.md

Add NVIDIA stress loader selection and DCGM 4 support

2026-03-31 11:15:15 +03:00

Description

No description provided

78 MiB

Languages

Go 83.8%

Shell 12.3%

C 3.8%

Dockerfile 0.1%