[Feature] performance problem
Open
Nobody has claimed this yet.
enhancement
- Dominant language
- Python
- Stars
- 624
- Forks
- 89
- PR merge metrics
- No merged PRs in 30d
Description
Is your feature request related to a problem? Please describe.
非常赞赏学长们的工作！我有一个小小的问题注意到readme里有一个吞吐和显存占用的表格。BMtrain显著优于Deepspeed- megaton，我好奇这其中的优化主要来源于什么地方呢。同样的逻辑，为什么我们能够支持更多的bach size，吞吐更高？是否也有显卡配置的原因呢（sxm的机器是不是会因为高带宽抹去这样的差距）。
我觉得做到这样的优化绝对是系统顶会级别的工作，学长们有兴趣分析这其中的优化点并总结成文章投稿吗。
其实我非常希望使用BMtrain的框架，但是只看到其中的好，不知道为什么好，心里就不踏实。
Describe the solution you'd like
同上
Describe alternatives you've considered
No response
Additional context
No response
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start with the benchmark table in the README and compare BMTrain with DeepSpeed-Megatron, including the reported batch size, throughput, and memory usage. Investigate whether GPU configuration, especially SXM systems, affects the comparison. Done means the optimization sources and hardware effects are analyzed and summarized in an article or equivalent documentation.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- python
- Domain
- machine-learning, performance
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100