Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

MultiThread + MultiQP in DistTensorClient #22

Open
eedalong opened this issue May 25, 2022 · 0 comments
Open

MultiThread + MultiQP in DistTensorClient #22

eedalong opened this issue May 25, 2022 · 0 comments
Assignees
Labels
Doing enhancement New feature or request

Comments

@eedalong
Copy link
Member

eedalong commented May 25, 2022

由于NIC本身具有多个处理器(Process Unit, PU),使用RC模式通信时,每一个QP绑定一个PU处理,为了避免性能卡在PU的处理性能上,我们在实现中使用了单线程多QP的方式来充分利用CPU能力。但同时目前发现单Client进行特征聚合时网络带宽只能用到10.5GB左右,距离12G仍然有2GB的距离,此时的瓶颈主要在于Client的CPU上了(具体可以查看测试脚本)。为此我们需要实现MultiThread, MultiQP的模式,避免单CPU瓶颈。同时Thread的个数需要暴露给用户进行设置,默认是1,一般最大设置到2应该就能够完全打满网络。

@eedalong eedalong assigned eedalong and Aiemu and unassigned eedalong and Aiemu May 25, 2022
@eedalong eedalong added the enhancement New feature or request label May 25, 2022
@eedalong eedalong added the Doing label May 25, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Doing enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

3 participants