Skip to content

boo1245/gpustat

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

68 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

gpustat

pypi Build Status license

Just less than nvidia-smi?

Screenshot: gpustat -cp

Usage

$ gpustat

Options:

  • --no-color : Suppress color (by default, color is enabled)
  • -u, --show-user : Display username of the process owner
  • -c, --show-cmd : Display the process name
  • -p, --show-pid : Display PID of the process
  • --json : JSON Output (Experimental, #10)

Tips

  • To periodically watch, try watch --color -n1.0 gpustat (built-in watch support will be added soon).
  • Running nvidia-smi daemon (root privilege required) will make the query much faster.
  • The GPU ID (index) shown by gpustat (and nvidia-smi) is PCI BUS ID, while CUDA differently assigns the fastest GPU with the lowest ID by default. Therefore, in order to make CUDA and gpustat use same GPU index, configure the CUDA_DEVICE_ORDER environment variable to PCI_BUS_ID (before setting CUDA_VISIBLE_DEVICES for your CUDA program): export CUDA_DEVICE_ORDER=PCI_BUS_ID

Quick Installation

Install from PyPI:

sudo pip install gpustat

To install the latest version (master branch) via pip:

pip install git+https://github.com/wookayin/gpustat.git@master

If you don't have root privilege, you can try pip install --user as well. Please note that from v0.4, gpustat.py is no more a zero-dependency executable.

sudo wget https://git.io/gpustat.py -O /usr/local/bin/gpustat && sudo chmod +x /usr/local/bin/gpustat

License

MIT License

About

📊 A simple command-line script (wrapper for nvidia-smi) for querying and monitoring GPU status

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Python 100.0%