high-performance SingleThreadedWorkQueue #35086
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Thanks for your contribution! |
Aurelius84
approved these changes
Aug 23, 2021
|
|
||
| EventCount(const EventCount&) = delete; | ||
|
|
||
| void operator=(const EventCount&) = delete; |
There was a problem hiding this comment.
Not important，我们其实是有DISABLE_COPY_AND_ASSIGN宏的
raindrops2sea
approved these changes
Aug 24, 2021
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
PR types
New features
PR changes
Others
Describe
基于Eigen的实现，改进实现细节，实现一个high-performance SingleThreadedWorkQueue。
对于Eigen原本的实现，修改点：
1）修改内存对齐和内存分配，进一步提升性能。
2）修改EventCount接口，提升易用性。
3）新增WaitQueueEmpty接口，便于用户等待task完成而无需自己追踪task。
4）替换了Eigen自定义宏为C++标准库宏和函数。
5）后续将要做的修改：将RunQueue的std::mutex替换为spinlock，以便提升性能
6）后续将要做的修改：将ThreadPool的多线程分支大改，包括spin等待逻辑/spin的条件，以便提升性能。
性能测试结论：
性能好于Paddle原来使用的ThreadPool、TFRT的SingleThreadedWorkQueue。
测试方法：
1）将PTB模型的OP计算图dump成文件，在测试程序中还原计算图。
2）使用一段纯CPU计算（for循环计数）模拟OP执行。
3）按照拓扑排序执行计算图，通过AddTask方法将算子提交到SingleThreadedWorkQueue。
4）执行2000个batch，每个batch执行一遍计算图，模拟训练过程。