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.

Sign up for GitHub

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

Jump to bottom

Vectorize ScaleCUDAKernel in DistributedFusedLamb #40011

Merged
merged 1 commit into from
Mar 2, 2022
Merged

Vectorize ScaleCUDAKernel in DistributedFusedLamb #40011

merged 1 commit into from
Mar 2, 2022

Conversation

sneaxiy
Copy link
Collaborator

@sneaxiy sneaxiy commented Mar 1, 2022 •
edited
Loading

PR types

Performance optimization

PR changes

OPs

Describe

对DistributedFusedLamb里的ScaleCUDAKernel做向量化优化，VecSize = 8 (FP16)。

单个batch的时间：

时间(ms) 性能数据 提升百分比
优化前（657次平均） 1268765.3 -
优化后 （657次平均） 750867.1 +40.8%

注：这个Kernel在单机8卡时用不到，主要是单机单卡才会用到。

@paddle-bot-old
Copy link

paddle-bot-old bot commented Mar 1, 2022

Thanks for your contribution!
Please wait for the result of CI firstly. See Paddle CI Manual for details.

@sneaxiy sneaxiy requested a review from limin2021 March 1, 2022 06:36
limin2021
limin2021 approved these changes Mar 2, 2022
View reviewed changes
Copy link
Contributor

@limin2021 limin2021 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM.

@sneaxiy sneaxiy merged commit 2e6548a into PaddlePaddle:develop Mar 2, 2022
@sneaxiy sneaxiy deleted the vec_scale_kernel branch March 2, 2022 10:58
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@limin2021 limin2021 limin2021 approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@sneaxiy @limin2021