Skip to main content

A tool for enriching the output of nvidia-smi

Project description

nvidia-htop

A tool for enriching the output of nvidia-smi.

CI PyPI version

Install

pip3 install nvidia-htop

Yes, this tool has been on PyPi since 2021! Enjoy the super-easy way to install it.

Usage

nvidia-htop.py [-l [length]]
  print GPU utilization with usernames and CPU stats for each GPU-utilizing process

  -l|--command-length [length]     Print longer part of the commandline. If `length'
                                   is provided, use it as the commandline length,
                                   otherwise print first 100 characters.
  -c|--color                       Colorize the output (green - free GPU, yellow -
                                   moderately used GPU, red - fully used GPU)
  -u|--user USER[,USER]            Limit the list of processes to selected users
                                   (comma-separated).
  -i|--id ID[,ID[,ID]]             Limit the command to selected GPU IDs (comma-separated).

Note: for backward compatibility, nvidia-smi | nvidia-htop.py [-l [length]] is also supported.

Note: running inside a container (docker, singularity, ...), nvidia-smi can only see processes running in the container.

Note: To periodically check the output of nvidia-htop, use the watch utility: watch nvidia-htop.py. To get colored output, you have to pass option -c to both watch and nvidia-htop, e.g. watch -c nvidia-htop.py -c.

Example output

$ nvidia-htop.py -l
Mon May 21 15:06:35 2018
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 390.25                 Driver Version: 390.25                    |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  GeForce GTX 108...  Off  | 00000000:04:00.0 Off |                  N/A |
| 53%   75C    P2   174W / 250W |  10807MiB / 11178MiB |     97%      Default |
+-------------------------------+----------------------+----------------------+
|   1  GeForce GTX 108...  Off  | 00000000:05:00.0 Off |                  N/A |
| 66%   82C    P2   220W / 250W |  10783MiB / 11178MiB |    100%      Default |
+-------------------------------+----------------------+----------------------+
|   2  GeForce GTX 108...  Off  | 00000000:08:00.0 Off |                  N/A |
| 45%   67C    P2    85W / 250W |  10793MiB / 11178MiB |     51%      Default |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
|  GPU   PID     USER    GPU MEM  %MEM  %CPU  COMMAND                                                                                               |
|    0  1032 anonymou   10781MiB   308   3.7  python train_image_classifier.py --train_dir=/mnt/xxxxxxxx/xxxxxxxx/xxxxxxxx/xxxxxxx/xxxxxxxxxxxxxxx  |
|    1 11021 cannotte   10765MiB   114   1.5  python3 ./train.py --flagfile /xxxxxxxx/xxxxxxxx/xxxxxxxx/xxxxxxxxx/xx/xxxxxxxxxxxxxxx                |
|    2 25544 nevermin   10775MiB   108   2.0  python -m xxxxxxxxxxxxxxxxxxxxxxxxxxxxx                                                               |
+-----------------------------------------------------------------------------+

Screenshot with output coloring

Screenshot

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

nvidia-htop-1.2.0.tar.gz (7.5 kB view hashes)

Uploaded Source

Built Distribution

nvidia_htop-1.2.0-py3-none-any.whl (7.1 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page