Dell iDRAC Telemetry Reference Guide
GPU Statistics
GPU statistics provide detailed error tracking and memory health insights for GPU components. These metrics focus on single-bit and double-bit error counters across various GPU memory regions, including frame buffer, L1/L2 cache, register files, and texture units. The data includes both real-time and cumulative error counts, and the number of memory pages retired due to errors. These statistics are essential for diagnosing hardware reliability, detecting memory faults, and maintaining GPU stability in high-performance computing environments.
| Metric Ids | Description | Metric Type | Units | Sensing Interval in seconds |
|---|---|---|---|---|
| SBECounterFB | Single-bit errors from Frame Buffer in Frame Buffer memory | Counter | Not Applicable | 600 |
| SBECounterFBL2Cache | No of Single-bit errors from L2 cache in Frame Buffer memory | Counter | Not Applicable | 600 |
| SBECounterGRRF | Single-bit errors from SM register file (RF) in Graphic Device (GR) memory | Counter | Not Applicable | 600 |
| SBERetiredPages | Number of pages dynamically retired because of single-bit errors in Frame Buffer memory | Counter | Not Applicable | 600 |
| SBECounterGRTex | Single-bit errors from texture units (TEX) in Graphic Device (GR) memory | Counter | Not Applicable | 600 |
| SBECounterGRL1Cache | Single-bit errors from L1 cache in Graphic Device (GR) memory | Counter | Not Applicable | 600 |
| CumulativeDBECounterFB | Cumulative double-bit error counter for Frame Buffer (FB) | Counter | Not Applicable | 600 |
| CumulativeSBECounterFB | Cumulative single-bit error counter for Frame Buffer (FB) | Counter | Not Applicable | 600 |
| CumulativeSBECounterGR | Cumulative single-bit error counter for Graphic Device (GR) | Counter | Not Applicable | 600 |
| CumulativeDBECounterGR | Cumulative double-bit error counter for Graphic Device (GR) | Counter | Not Applicable | 600 |
| DBECounterFB | Double-bit errors from Frame Buffer in Frame Buffer memory | Counter | Not Applicable | 600 |
| DBECounterGRL1Cache | Double-bit errors from L1 cache in Graphic Device (GR) memory | Counter | Not Applicable | 600 |
| DBECounterFBL2Cache | Double-bit errors from L2 cache in Frame Buffer memory | Counter | Not Applicable | 600 |
| DBECounterGRTex | Double-bit errors from texture units (TEX) in Graphic Device (GR) memory | Counter | Not Applicable | 600 |
| DBERetiredPages | Number of pages dynamically retired because of double-bit errors in Frame Buffer memory | Counter | Not Applicable | 600 |
| DBECounterGRRF | Double-bit errors from SM register file (RF) in Graphic Device (GR) memory | Counter | Not Applicable | 600 |
Please provide ratings (1-5 stars).
Please provide ratings (1-5 stars).
Please provide ratings (1-5 stars).
Please select whether the article was helpful or not.
Comments cannot contain these special characters: <>()\