New poll_barrier as alternative to mpi.barrier #7
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
Wentzell
approved these changes
May 3, 2021
MPI_Barrier produces 100% CPU load on waiting mpi ranks. As an alternative one can combine a non blocking MPI_IBarrier with MPI_Test each X milliseconds to only check every so often if all ranks reached the barrier. Already X=10msec reduces the CPU load below 1% load, making it possible to use the rank for other tasks in MPI / OpenMP hybrid settings, or allowing more efficient thermal use of the CPU. Thanks to A. Pataki for helping! Changes: - implement function barrier(msec) to check only each msec the barrier - is now the new default barrier function with msec=1 as default - setting 0 msec as polling freq. the function calls normal MPI_Barrier - add test mpi_barrier.cpp
|
Thank you @the-hampel for adding this! |
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Outdated suggestions cannot be applied.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.
MPI_Barrier produces 100% CPU load on waiting mpi ranks. As an
alternative one can combine a non blocking MPI_IBarrier with MPI_Test
each X milliseconds to only check every so often if all ranks reached
the barrier. Already X=100msec reduces the CPU load below 1% load,
making it possible to use the rank for other tasks in MPI / OpenMP
hybrid settings, or allowing more efficient thermal use of the CPU.
Thanks to A. Pataki for helping!
Changes:
barrier