Skip to content

green-hand-shandong/SynerFuse

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

13 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

SynerFuse

OpenSSF Best Practices

Latest News

  • [2025/6]SynerFuse publishes framework of Release 1.0, which can provide heterogeneous pipeline parallelism and heterogeneous data parallelism capabilities.
  • [2025/5] China Mobile has established the repository for SynerFuse and obtained the OpenSSF certification badge.

SynerFuse Overview

Currently, the “resource wall” between different GPUs makes it difficult to build one heterogeneous resource pool for Large-scale models training. Heterogeneous distributed training becomes a pressing challenge for the industry to solve. We brought up a cross-architecture unified heterogeneous training framework SynerFuse to deal with the problem.

SynerFuse enables multiple LLMs deployed and trained on multiple types of GPUs. The Inhomogeneous Task Distribution(ITD) algorithm for heterogeneous training task splitting is innovatively proposed, which supports heterogeneous data parallelism(DP) and heterogeneous pipeline parallelism(PP), and realizes the adaptive adjustment of parameters such as size and number of micro batches in DP, stages and layers in PP on heterogeneous GPUs.

We’ve verified our capability on Nvidia and other 4 types of GPUs. The acceleration ratio reached 95%, loss converges to 1.8 and PPL curve converged normally.

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 97.7%
  • Shell 1.3%
  • C++ 0.8%
  • C 0.1%
  • HTML 0.1%
  • Dockerfile 0.0%