lindahl at keyresearch.com
Sat May 4 10:56:15 PDT 2002
On Sat, May 04, 2002 at 10:10:16AM +0200, joachim wrote:
> > If a benchmark measures process-to-process bandwidth and latency
> > between only 2 processes, it matters whether or not they are on the
> > same node (shmem) or not. It's not a matter of round robin vs. groups.
> I need to refer to the disclaimer again:
Joachim, this is why I normally ignore your postings instead of
answering them. Disclaiming something which is wrong is no excuse.
More information about the Beowulf