Get detailed insights into congestion on the High Bandwidth Memory (HBM), such as the devices on which congestion has occurred, the timestamps, and when the device returned to its normal state.
High Bandwidth Memory (HBM) congestion detection is a Priority Flow Control (PFC) congestion monitoring functionality that
-
helps monitor buffer occupancy across the Shared Memory System (SMS) and High-Bandwidth Memory (HBM)
-
detects when Virtual Output Queue (VOQ) spillover into HBM indicates congestion, and
-
records time-stamped congestion events and history for post-event analysis and troubleshooting.
| Feature Name |
Release Information |
Feature Description |
|---|---|---|
| Detect High Bandwidth Memory Congestion |
Release 25.4.1 |
Introduced in this release on: Fixed Systems (8010 [ASIC: A100])(select variants only*) *This feature is supported on:
|
| Detect High Bandwidth Memory Congestion |
Release 25.1.1 |
Introduced in this release on: Fixed Systems (8700 [ASIC: K100], 8010 [ASIC: A100])(select variants only*) This feature is supported on:
|
| Detect High Bandwidth Memory Congestion |
Release 24.4.1 |
Introduced in this release on: Fixed Systems (8200 [ASIC: P100], 8700 [ASIC: P100])(select variants only*); Modular Systems (8800 [LC ASIC: P100])(select variants only*) *This feature is supported on:
|
| Detect High Bandwidth Memory Congestion |
Release 7.5.3 |
We provide detailed insights into congestion on the High Bandwidth Memory (HBM), such as the devices on which congestion has occurred, the time stamps, and when the device returned to its normal state. With such details, you can investigate the cause of congestion and identify the source ports causing congestion for future preventive actions. You must configure PFC in the buffer-extended mode for this option. The feature introduces the following to enable the option to detect HBM congestion:
It also introduces the following to view the congestion and memory usage details:
|