Skip to content

PluralisResearch/AsyncMesh

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

This is the official implementation of the paper titled AsyncMesh: Fully Asynchronous Optimization for Data and Pipeline Parallelism.

Instructions

First install requirements.txt and run the bash script run.bash. This script assumes an instance with at least 8 GPUs and runs our method for the base model on the 4 x 2 mesh. Tested on PyTorch 2.5.1, CUDA 12.6, and Python 3.12.

Acknowledgements

License

Copyright © Pluralis Research. All rights reserved.

This project is licensed under the MIT License. See the LICENSE file for details.

Citation

@article{ajanthan2026asyncmesh,
  title={AsyncMesh: Fully Asynchronous Optimization for Data and Pipeline Parallelism},
  author={Ajanthan, Thalaiyasingam and Ramasinghe, Sameera and Avraham, Gil and Mohaghegh Dolatabadi, Hadi and P Hewa Koneputugodage, Chamin and Shevchenko, Violetta and Zuo, Yan and Long, Alexander},
  journal={arXiv:2601.22442},
  year={2026}
}

About

AsyncMesh Implementation

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors