Add benchmark fan duty cycle summary to report

This commit is contained in:
Mikhail Chusavitin
2026-04-14 10:24:02 +03:00
parent 88b5e0edf2
commit 8fc986c933
7 changed files with 247 additions and 14 deletions

View File

@@ -13,15 +13,18 @@ import (
// GPUMetricRow is one telemetry sample from nvidia-smi during a stress test.
type GPUMetricRow struct {
Stage string `json:"stage,omitempty"`
ElapsedSec float64 `json:"elapsed_sec"`
GPUIndex int `json:"index"`
TempC float64 `json:"temp_c"`
UsagePct float64 `json:"usage_pct"`
MemUsagePct float64 `json:"mem_usage_pct"`
PowerW float64 `json:"power_w"`
ClockMHz float64 `json:"clock_mhz"`
MemClockMHz float64 `json:"mem_clock_mhz"`
Stage string `json:"stage,omitempty"`
ElapsedSec float64 `json:"elapsed_sec"`
GPUIndex int `json:"index"`
TempC float64 `json:"temp_c"`
UsagePct float64 `json:"usage_pct"`
MemUsagePct float64 `json:"mem_usage_pct"`
PowerW float64 `json:"power_w"`
ClockMHz float64 `json:"clock_mhz"`
MemClockMHz float64 `json:"mem_clock_mhz"`
FanAvgRPM float64 `json:"fan_avg_rpm,omitempty"`
FanDutyCyclePct float64 `json:"fan_duty_cycle_pct,omitempty"`
FanDutyCycleAvailable bool `json:"fan_duty_cycle_available,omitempty"`
}
// sampleGPUMetrics runs nvidia-smi once and returns current metrics for each GPU.
@@ -142,10 +145,14 @@ func sampleAMDGPUMetrics() ([]GPUMetricRow, error) {
// WriteGPUMetricsCSV writes collected rows as a CSV file.
func WriteGPUMetricsCSV(path string, rows []GPUMetricRow) error {
var b bytes.Buffer
b.WriteString("stage,elapsed_sec,gpu_index,temperature_c,usage_pct,mem_usage_pct,power_w,clock_mhz,mem_clock_mhz\n")
b.WriteString("stage,elapsed_sec,gpu_index,temperature_c,usage_pct,mem_usage_pct,power_w,clock_mhz,mem_clock_mhz,fan_avg_rpm,fan_duty_cycle_pct,fan_duty_cycle_available\n")
for _, r := range rows {
fmt.Fprintf(&b, "%s,%.1f,%d,%.1f,%.1f,%.1f,%.1f,%.0f,%.0f\n",
strconv.Quote(strings.TrimSpace(r.Stage)), r.ElapsedSec, r.GPUIndex, r.TempC, r.UsagePct, r.MemUsagePct, r.PowerW, r.ClockMHz, r.MemClockMHz)
dutyAvail := 0
if r.FanDutyCycleAvailable {
dutyAvail = 1
}
fmt.Fprintf(&b, "%s,%.1f,%d,%.1f,%.1f,%.1f,%.1f,%.0f,%.0f,%.0f,%.1f,%d\n",
strconv.Quote(strings.TrimSpace(r.Stage)), r.ElapsedSec, r.GPUIndex, r.TempC, r.UsagePct, r.MemUsagePct, r.PowerW, r.ClockMHz, r.MemClockMHz, r.FanAvgRPM, r.FanDutyCyclePct, dutyAvail)
}
return os.WriteFile(path, b.Bytes(), 0644)
}