Applicable to Entuity v23.0 P03 upwards.
Incidents and events:
- Enable GPU Uncorrected ECC Error Rate
- GPU Corrected ECC Error Rate
- GPU High Fan Speed (%)
-
GPU High Temperature
- GPU High Temperature Critical
- GPU High Temperature Warning
- GPU High Memory Used (%)
- GPU High Power Usage (% of limit)
-
GPU High Utilization (%)
- GPU High Utilization (%)
- GPU High Utilization Warning (%)
- GPU Instance High FB Memory Used (%)
- GPU Link Down Events
- GPU Link Error Rate
-
GPU Memory High Temperature
Thresholds:
The following warning thresholds are applicable to monitoring GPUs in Entuity. For more information on thresholds in Entuity, see the following articles:
- How do I enable event thresholds (static thresholds and dynamic thresholds)?
- Thresholds dashboard
- Thresholds Summary dashlet
| StormWorks name | Display name | Attribute | Threshold type | Default values | Range | Applicable to which OS? |
|---|---|---|---|---|---|---|
| gpuTotalCorrectedECCErrors | GPU Corrected ECC Error Rate | ECC Errors | Yellow | 0 | 0-100 | Windows and Linux |
| gpuFanSpeed | GPU High Fan Speed (%) | Fan Speed% | Red | 90 | 0-100 | Windows and Linux |
| gpuTemperature | GPU High Temperature Critical (C) | GPU Temperature | Red | 92 | 0-100 | Windows and Linux |
| gpuTemperature | GPU High Temperature Warning (C) | GPU Temperature | Yellow | 85 | 0-100 | Windows and Linux |
| gpuMemoryTemperature | GPU Memory High Temperature (C) | GPU Memory Temperature | Red | 90 | 0-100 | Windows and Linux |
| gpuTotalUncorrectedECCErrors | Enable GPU Uncorrected ECC Error Events | ECC Errors | Red | 1 | 1/2, enabled/disabled | Windows and Linux |
| gpuMemoryUsed vs gpuMemoryTotal | GPU High Memory Used (%) | GPU Memory Used (%) | Red | 90 | 0-100 | Windows and Linux |
| gpuPowerUsage vs gpuPowerLimit | GPU High Power Usage (% of limit) | GPU Power Usage (%) | Red | 95 | 0-100 | Windows and Linux |
| gpuUtilization | GPU High Utilization (%) | GPU Utilization (%) | Red | 90 | 0-100 | Windows and Linux |
| gpuUtilization | GPU High Utilization Warning (%) | GPU Utilization (%) | Yellow | 80 | 0-100 | Windows and Linux |
| giFBUsed vs giFBTotal | GPU Instance High FB Memory Used (%) | GPU Instance Memory Used (%) | Red | 90 | 0-100 | Windows and Linux |
| glLinksDown | GPU Link Down Events | GPU Link Down | Red | 1 | 1/2, enabled/disabled | Windows and Linux |
| alCRCFlitErrors, alCRCDataErrors, alReplayErrors and alRecoveryErrors | GPU Link Error Rate | GPU Link Error Rate | Yellow | 0 | 0-100 | Windows and Linux |
Comments
0 comments
Please sign in to leave a comment.