modelscope / modelscope/DiffSynth-Studio

nproc_per_node issue

Open
#678 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
13.1k
Forks
1.3k
Avg merge
13h 12m
Merged PRs (30d)
45

Description

torchrun --standalone --nproc_per_node=1 examples/wanvideo/acceleration/unified_sequence_parallel.py
why is much slower than normally run 'python unified_sequence_parallel.py'
which leads '--nproc_per_node=2 examples/wanvideo/acceleration/unified_sequence_parallel.py' gain small speedup than normally run 'python unified_sequence_parallel.py'.

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by reproducing the two commands against examples/wanvideo/acceleration/unified_sequence_parallel.py and compare their timings with nproc_per_node set to 1 and 2. Read that entry point and the torchrun-related setup it uses; done means the performance difference is explained and any required change is verified with comparable runs.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
distributed-systems, machine-learning, performance
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
30/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.