reanimator/bee
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 1 Wiki Activity

Adjust burn GPU ramp timing by profile

Browse Source
This commit is contained in:
Michael Chus
2026-04-12 15:58:30 +03:00
parent 525ed8b8fc
commit e16d0f34b5
3 changed files with 156 additions and 24 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
audit/internal/webui/pages.go
View File

@@ -1342,11 +1342,11 @@ func renderValidate(opts HandlerOptions) string {
<div class="grid3">
` + renderSATCard("nvidia", "NVIDIA GPU", "runNvidiaValidateSet('nvidia')", "", renderValidateCardBody(
inv.NVIDIA,
`Runs NVIDIA diagnostics and board inventory checks.`,
`<code>nvidia-smi</code>, <code>dmidecode</code>, <code>dcgmi diag</code>`,
`Level 2 in Validate, Level 3 in Stress. Runs one GPU at a time on the selected NVIDIA GPUs.`,
)) +
inv.NVIDIA,
`Runs NVIDIA diagnostics and board inventory checks.`,
`<code>nvidia-smi</code>, <code>dmidecode</code>, <code>dcgmi diag</code>`,
`Level 2 in Validate, Level 3 in Stress. Runs one GPU at a time on the selected NVIDIA GPUs.`,
)) +
`<div id="sat-card-nvidia-targeted-stress">` +
renderSATCard("nvidia-targeted-stress", "NVIDIA GPU Targeted Stress", "runNvidiaValidateSet('nvidia-targeted-stress')", "", renderValidateCardBody(
inv.NVIDIA,
@@ -2373,7 +2373,7 @@ func renderBurn() string {
<p id="burn-selection-note" style="font-size:12px;color:var(--muted);margin:10px 0 0">Select at least one NVIDIA GPU to enable NVIDIA burn recipes.</p>
<label class="cb-row" style="margin-top:10px">
<input type="checkbox" id="burn-stagger-nvidia">
<span>Ramp selected NVIDIA GPUs one by one before full-load hold. Uses a 3-minute stabilization window per GPU, then keeps all selected GPUs under load for the chosen Burn Profile duration.</span>
<span>Ramp selected NVIDIA GPUs one by one before the full-load hold. Smoke: +2 min per GPU, then 5 min with all selected GPUs under load. Acceptance: +10 min per GPU, then at least 1 hour with all selected GPUs under load. Overnight: +1 hour per GPU, then at least 1 hour with all selected GPUs under load, capped at 10 hours total.</span>
</label>
</div>
</div>
@@ -3108,7 +3108,6 @@ usbRefresh();
</script>`
}
func renderNvidiaSelfHealInline() string {
return `<p style="font-size:13px;color:var(--muted);margin-bottom:12px">Inspect NVIDIA GPU health, restart the bee-nvidia driver service, and issue a per-GPU reset when the driver reports reset required.</p>
<div style="display:flex;gap:8px;flex-wrap:wrap;margin-bottom:12px">