1

nvidia-smi is showing as

+-----------------------------------------------------------------------------+
| NVIDIA-SMI 440.33.01    Driver Version: 440.33.01    CUDA Version: 10.2     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  Quadro K620         On   | 00000000:02:00.0  On |                  N/A |
| 65%   74C    P0    20W /  30W |    758MiB /  1994MiB |     98%      Default |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID   Type   Process name                             Usage      |
|=============================================================================|
|    0      1314      G   /usr/lib/xorg/Xorg                            31MiB |
|    0      1927      G   /usr/lib/xorg/Xorg                           187MiB |
|    0      2139      G   /usr/bin/gnome-shell                         248MiB |
|    0      4055      G   ...AAAAAAAAAAAACAAAAAAAAAA= --shared-files    85MiB |
|    0      5824      G   /opt/teamviewer/tv_bin/TeamViewer              8MiB |
|    0     14613      C   gmx                                           73MiB |
|    0     21803      G   /usr/lib/rstudio/bin/rstudio                  59MiB |
+-----------------------------------------------------------------------------+

As you can see GPU util is showing 98%

But RAM is still free - 758MiB / 1994MiB

What does exactly GPU-Util mean? Can I load one more process into the GPU now?

Especially I wish to load a molecular dynamics simulation to GPU. (gmx)

1 Answer 1

1

GPU-Util is the percentage of time, over the last sample period, during which at least one kernel was running on the GPU.

98% means that your GPU is being used nearly all the time, probably by gmx, so there might not be any spare capacity to run another compute task.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.