This repository was archived by the owner on Nov 17, 2023. It is now read-only.
enabling multithreading in broadcast_reduce#9444
Merged
piiswrong merged 1 commit intoapache:masterfrom Jan 15, 2018
Merged
Conversation
Contributor
|
LGTM. FYI, the speed of broadcast reduce can be further increased (5x ~ 10x factor) if it uses LaunchEx I only did it for binary_broadcast. You are welcome to contribute to the other parts if you want |
CodingCat
pushed a commit
to CodingCat/mxnet
that referenced
this pull request
Jan 16, 2018
Contributor
Author
|
Will follow that pattern in order to make things consistent. |
larroy
pushed a commit
to larroy/mxnet
that referenced
this pull request
Jan 18, 2018
yuxiangw
pushed a commit
to yuxiangw/incubator-mxnet
that referenced
this pull request
Jan 25, 2018
rahul003
pushed a commit
to rahul003/mxnet
that referenced
this pull request
Jun 4, 2018
zheng-da
pushed a commit
to zheng-da/incubator-mxnet
that referenced
this pull request
Jun 28, 2018
This file contains hidden or 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
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Description
Add multithreading on CPU for the class of broadcast_reduce operators. For unknown reasons, this class of operators does not use any internal threading so far and therefore was observed to become a serious runtime bottleneck in an application. Threading is done on the level of sequences to be reduced, not within a single reduce sequence. This pattern is in line with the one that we are already doing in elemwise_binary_broadcast_op.h (where we achieve threading when starting the binary_broadcast_kernel).
With this change, this class of operators shows good threading whenever we reduce over multiple sequences and matches the runtime characteristics of elemwise_binary_broadcast.
Checklist
Essentials
make lint)