[Beowulf] China to eclipse Titan with 48,000 Intel MICs?
prentice.bisbal at rutgers.edu
Thu Jun 6 07:26:49 PDT 2013
On 06/06/2013 04:40 AM, Hearns, John wrote:
> > No, that's indeed what happens on most supers I've seen. Even if some
> > codes are capable of running to the (almost) full scale of the
> > machine, such large runs are quite marginal. They often take place
> > before the system enters production, to demonstrate its capabilities,
> Well said.
> > > Along similar lines, has Google or Amazon ever published their
> batch job
> > > capacity (it must be in petaflops...)
> > That means high latencies and grumpy MPI.
> Grumpy MPI. Now there's a name for a new open source project, if I
> have ever seen one.
You spelled that wrong. It's spelled GruMPI®
> Let's start it now - a new MPI implementation which is intended for
> clouds and geographically
> separated machines with high latencies between them. The MPI
> implementation will cope with
> Network interruptions of a long duration, but will of course produce
> copious grumpy error messages.
> And we can have mailing lists which are famed across the Internet for
> their spikiness and flame wars.
> I vote we invite Victor Meldrew to be the project's new figurehead.
I vote for Grumpy Cat (or should I say GruMPI® cat?) :
More information about the Beowulf