[petsc-dev] PETSc OpenMP benchmarking

[Jed Brown]

On Sat, Mar 17, 2012 at 10:52, Gerard Gorman <g.gorman at imperial.ac.uk>wrote:

> Ah! One of those files *was* from a different run. Here are the correct
> files for comparison (this time compiled with -O3). Sorry for the
> confusion.
>

This looks like a _tiny_ problem size. Like a few thousand degrees of
freedom total. That is small enough that you even be seeing things like
cache line write contention. Also, how is this matrix ordered? How much
difference do you see if you reorder it (e.g. with RCM) before the solve?
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20120317/f6b887c4/attachment.html>