Refactor validate modes, fix benchmark report and IPMI power

- Replace diag level 1-4 dropdown with Validate/Stress radio buttons
- Validate: dcgmi L2, 60s CPU, 256MB/1p memtester, SMART short
- Stress: dcgmi L3 + targeted_stress in Run All, 30min CPU, 1GB/3p memtester, SMART long/NVMe extended
- Parallel GPU mode: spawn single task for all GPUs instead of splitting per model
- Benchmark table: per-GPU columns for sequential runs, server-wide column for parallel
- Benchmark report converted to Markdown with server model, GPU model, version in header; only steady-state charts
- Fix IPMI power parsing in benchmark (was looking for 'Current Power', correct field is 'Instantaneous power reading')

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
2026-04-08 00:42:12 +03:00
parent dd26e03b2d
commit b2f8626fee
12 changed files with 332 additions and 164 deletions

View File

@@ -326,8 +326,8 @@ func (s *System) RunNvidiaBenchmark(ctx context.Context, baseDir string, opts Nv
}
report := renderBenchmarkReportWithCharts(result, loadBenchmarkReportCharts(runDir, selected))
if err := os.WriteFile(filepath.Join(runDir, "report.txt"), []byte(report), 0644); err != nil {
return "", fmt.Errorf("write report.txt: %w", err)
if err := os.WriteFile(filepath.Join(runDir, "report.md"), []byte(report), 0644); err != nil {
return "", fmt.Errorf("write report.md: %w", err)
}
summary := renderBenchmarkSummary(result)
@@ -1183,18 +1183,8 @@ func queryIPMIServerPowerW() (float64, error) {
if err != nil {
return 0, fmt.Errorf("ipmitool dcmi power reading: %w", err)
}
for _, line := range strings.Split(string(out), "\n") {
if strings.Contains(line, "Current Power") {
parts := strings.SplitN(line, ":", 2)
if len(parts) == 2 {
val := strings.TrimSpace(strings.TrimSuffix(strings.TrimSpace(parts[1]), "Watts"))
val = strings.TrimSpace(val)
w, err := strconv.ParseFloat(val, 64)
if err == nil && w > 0 {
return w, nil
}
}
}
if w := parseDCMIPowerReading(string(out)); w > 0 {
return w, nil
}
return 0, fmt.Errorf("could not parse ipmitool dcmi power reading output")
}