Measuring GPU availability on Jetsons

The Xavier/Orin in our system has ~10 separate processes that launch CUDA kernels. We want to know if a GPU operation (CUDA kernel execution or Memory Copy) is blocked because of other GPU operations running from separate processes, since Jetsons do not support MPS or MIG.

I intend to use nvprof (on Xavier running CUDA 10) for this purpose. Is there a more efficient way to measure GPU availability?

To clarify: I am not looking to measure GPU utilization (tegrastats or the GPU load sysfs file)

Please try the tool:
How to get GPU usage by an specific application(PID) - #5 by AastaLLL

I will try it, once I upgrade to CUDA 11, thanks.

I have tried it on CUDA 10 on the Xavier, but it wasn’t able to attach to processes from inside a Docker container (nvprof was able to)

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.