[petsc-users] Variable wall times

[Jed Brown]

Phanisri Pradeep Pratapa <ppratapa at gatech.edu> writes:
> @Jed:
> There are no other jobs running on that node (and also no other jobs on the
> queue). So could the noise be due to load balance issue or something else
> in computing dot products and MatMults? Is this to be expected? If so, is
> the only work around for this is to report a statistics of wall time for
> the runs? (When I compare two different algorithms, the speed-up of one
> over the other varies significantly because of this issue.)

Try using only 32 processes.  Hyperthreads aren't full threads and you
could just be oversubscribed.  Also note that there are system processes
that can take time.  If the performance varies between runs, you have to
look at nondeterministic factors rather than deterministic factors like
partition-induced load balance.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 818 bytes
Desc: not available
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20160222/a588646a/attachment.pgp>