leefs
Results
2
issues of
leefs
We are using DCGM to monitor NVSwitch performance on a system equipped with NVIDIA H100 GPUs (HGX platform). While NVLink-related metrics such as DCGM_FI_PROF_NVLINK_TX_BYTES and DCGM_FI_PROF_NVLINK_RX_BYTES report valid values, the...
### Ask your question I would like to clarify the granularity of the following NVLink metrics exposed by DCGM: DCGM_FI_PROF_NVLINK_RX_BYTES DCGM_FI_PROF_NVLINK_TX_BYTES Are these metrics reported per GPU (i.e., aggregated across...
question