Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Timers for local and remote sums in density and force #271

Merged
merged 32 commits into from
Apr 22, 2022

Conversation

conradtchan
Copy link
Collaborator

@conradtchan conradtchan commented Mar 16, 2022

Type of PR:
other

Description:
Add timers to density and force routines to compare the time spent computing sums on local particles vs remote particles.

Add column to timings: load bal - measures the MPI work load balance, where 100% is perfectly balanced. Defined as the mean CPU time of every task except for the slowest, divided by the slowest time.

Testing:
Verify timers are subsets of their parents

Did you run the bots? no

conradtchan and others added 30 commits March 10, 2022 17:40
…tom into memory-optimisation

# Conflicts:
#	build/Makefile
#	src/main/memory.F90
because maxprocs is needed to get the right size, which we don't want to set as a parameter

subroutines adjusted accordingly (replaced direction with remote_export)
amend previous commit
bugfix: missing nprocs import
amend previous commit
Change back to using threadprivate for xyzcache
update mailmap for david liptai and conrad chan
@danieljprice danieljprice marked this pull request as ready for review April 22, 2022 23:30
@danieljprice danieljprice merged commit f5dfb1b into danieljprice:master Apr 22, 2022
@conradtchan conradtchan deleted the mpi-timings branch August 23, 2022 00:01
s-neilson pushed a commit to s-neilson/phantom that referenced this pull request Mar 18, 2023
Timers for local and remote sums in density and force
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants