Parallelization via OpenMP support #6
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
olitheolix
reviewed
Nov 4, 2022
There was a problem hiding this comment.
@Abde5 It did not compile on Linux out of the box due to various
.../process.cpp:229:37: error: ‘pmeshs’ not specified in enclosing ‘parallel’
errors. However, I managed to make them disappear with the two suggested changes. CoACD now uses all cores during the decomposition phase of the algorithm.
Thank you for that PR 🚀
|
@olitheolix to be honest, the PR did not even compile on Windows 😅 I didn't PR with the latest commit in my local branch, thanks for fixing the problem, tho! |
Co-authored-by: Oliver Nagy <[email protected]>
|
Hi @Abde5 , |
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Outdated suggestions cannot be applied.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.
Dear Wei, Liu, Ling and Su,
Reading the Discussion chapter of your paper, we took the freedom to implement a parallelization of CoACD.
We did so by parallelizing part of the Compute problem (each CoACD iteration is perfectly parallelizable) and the merge (the cost matrix computing has been parallelized). We attempted to implement a more complex parallelization system with OpenMP tasks; however, it did not give good results compared to a basic parallelization of the for routines.
The Merge loop has been simplified in order to keep it a single loop.
With empirical tests, we have observed that we achieve speedups from 1.2 to 3.8 for the smaller concavity thresholds.
Please be free to try it out and I would be glad to answer any questions about the implementation.
This work is being contributed by Altheria Solutions.