On Fri, Nov 12, 2010 at 14:51, Rodrigo R. Paz <rodrigop at intec.unl.edu.ar>wrote:
> IMHO, the main problem here is the low memory bandwidth (FSB) in Xeon E5420 > nodes. This is well-known and fundamental (not dependent on the programming model). It appears that you are comparing OpenMP threading within a node to a single thread. The comparison we are more interested is with an equivalent number of MPI processes per node. Jed -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20101112/8553c489/attachment.html>
