search

NVIDIA / DCGM

NVIDIA Data Center GPU Manager (DCGM) is a project for gathering telemetry and measuring the health of NVIDIA GPUs
Apache License 2.0
393 stars 50 forks source link

How to get current memory usage in bytes from dmon? #177

Closed johnsushant closed 3 months ago

johnsushant commented 3 months ago

Is there a way to get current GPU memory usage in bytes from dmon? I would ideally expect a field like DCGM_FI_GPU_MEMORY_USAGE which provides this metric.

Is there any other recommended way to do this other than calling the NVML API directly?

glowkey commented 3 months ago

You can try DCGM_FI_DEV_FB_USED, it is in MB though. DCGM_FI_GPU_MEMORY_USAGE is not a valid field.