(DO NOT MERGE): Tpetra: CrsMatrix:: and CrsGraph:: copyAndPermute performance improvements #13682
+975
−85
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Outdated suggestions cannot be applied.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.
@trilinos/
Motivation
Clients have reported copyAndPermute as a large portion of the time spent in e.g. evaluating and scattering the Jacobian in the panzer application. This PR addresses that by profiling and rewriting code, and using kokkos parallel for/reduce.
Status
Serial speedups have been obtained by writing a batch/array version of getGlobalElement (which amortizes the 'if(contiguous)' and other if-stmts), rewriting replaceGlobalValuesImpl and getGlobalRowCopy.
This bottom-up approach led to a top-down investigation, leading to using Kokkos to parallelize the copyAndPermute main loop over rows of the source matrix.
Related Issues
#13587
13587
Stakeholder Feedback
Testing