ByteDance-Seed / ByteDance-Seed/Triton-distributed

which one is all reduce test file~~

Open
#12 1 comment 0 reactions 0 assignees View on GitHub
enhancement
Dominant language
Python
Stars
1.5k
Forks
172
PR merge metrics
No merged PRs in 30d

Description

### Describe the issue

hello, I can successfully test AG RS A2A kernel, thank you very much;
however, I want to test all reduce kernel, which python file I can use?
or how can I modify some python file to test all reduce kernel

![Image](https://github.com/user-attachments/assets/6f9f47c3-2ebc-4517-8454-7dd106a962e9)

### Environment details

- Python 3.9 (suggest using virtual environment)
- CUDA 12.4
- Torch 2.4.1
- Clang 19
a100 * 8

Contributor guide

Open the contributing guide

Research direction

No Python file, test, or entry point is named in the issue. Start by locating the repository's all-reduce kernel and its Python test or example entry point, then check whether the existing A2A test setup can exercise it with the stated Python, CUDA, Torch, and eight-A100 environment. Done means documenting the correct file or modification needed to run the all-reduce test.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
distributed-systems, testing
Issue type
Documentation
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.