use same MPB phase on all processes #578
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
Merged
Closed
bencbartlett
pushed a commit
to bencbartlett/meep
that referenced
this pull request
Sep 9, 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Outdated suggestions cannot be applied.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.
Hopefully fixes #568 … this looks like it has been a longstanding bug, I'm surprised we didn't notice it before now.
The basic problem I noticed is that MPB computes modes with a "random" phase, but it is a different random phase on every processor. Therefore, in any situation where we are using the same mode calculation distributed across multiple processors (either an eigenmode source that spans multiple chunks or a mode overlap integral that spans multiple chunks), the different processors will be doing the calculation with inconsistent phases.
(The fix uses the POSIX
rand_rfunction, which should be okay since I think we only run on POSIX platforms anyway? We don't require high-quality random numbers here, since we are only initializing to "random" values to make sure the starting vectors are not accidentally nearly orthogonal to the desired modes.)