[NCLUG] parallel processing users?

[Jeffrey D. Means]

I'll second this motion for a discussion on clusters.

:)
Jeff
On Fri, 2005-10-14 at 14:13 -0600, Evelyn Mitchell wrote:
> I'd be interested in a presentation (perhaps a panel discussion) on
> clustering (compute, grid, high availability).
> 
> Would anyone else be interested? 
> 
> Matt/Bob: would you be available to present?
> 
> I'd volunteer jafo for talking about High availability.
> 
> Evelyn Mitchell
> efm
> 
> * On 2005-10-14 13:05 Matt <rosing at peakfive.com> wrote:
> >  > John Gilmore wrote:
> >  > > Matt wrote:
> >  > > > Is there anyone lurking out there that does parallel processing in the
> >  > > > Northern Colorado area?  I'm trying to find like minded people, maybe
> >  > > > we can help each other.
> >  > >
> >  > > What exactly do you mean? Are you looking for people who are running dual 
> >  > > processors?
> >  > 
> > 
> > I do contract work for PNNL, a Dept of Energy lab in Washington
> > state.  They have a ~2000 processor(itanium) machine and it's used for
> > doing modeling in areas like chemistry and environmental cleanup.
> > Some people use just a few processors, some use the whole machine all
> > day, some use 200 processors for a week.  I do a lot of profiling and
> > looking at hardware counters and rewriting code to make things more
> > efficient. 
> > 
> > Anyway, performance is what it's all about.  I'm interested in finding
> > people that use parallel processing because a single processor can't
> > go fast enough.  Could be distributed jobs, or tightly communicating
> > models or data mining or computational biology or using specialized
> > processors or who knows what.  I'm open to anything.  It just seems
> > that now, since dual core processors are coming out, it might be the
> > case that making faster chips is harder than parallel chips, so
> > parallel processing might become more important.
> > 
> >  > Or are you fishing for people who work with "grid" queues of many
> >  > machines doing distributed job processing?  This would describe my
> >  > work situation.  I have been managing large compute pools for a number
> >  > of years.
> > 
> > Sure, if there's a desire to make things go faster. If you have a
> > hundred blades and everyone's happy with the turnaround time then I
> > guess it's not that interesting but if there are people that can see a
> > real benefit to making it go 10 times faster then it's interesting.
> > 
> >  > Or tightly coupled parallel processing such as Beowulf clusters?
> >  > Always been on my list to play with but never had the time to do so.
> > 
> > That's what I'm more familiar with, but I'd like to learn about new
> > things. 
> > 
> > Matt
> > 
> > 
> > 
> > _______________________________________________
> > NCLUG mailing list       NCLUG at nclug.org
> > 
> > To unsubscribe, subscribe, or modify 
> > your settings, go to: 
> > http://www.nclug.org/mailman/listinfo/nclug
> This email is: [ ] actionable [x] fyi [ ] social
> Response needed: [ ] yes [x] up to you [ ] no 
> Time-sensitive: [ ] immediate [x] soon [ ] none
> 
> 
-- 

Jeffrey D. Means                                   meaje at meanspc.com
Owner / CIO for MeansPC                       http://www.meanspc.com/
Custom Web Development For Your Needs.                 (970)308-1298

- Everything in moderation including moderation. - Unknown

My Public PGP Key ID is: 0x81F00126
and available via:  
http://pgp.mit.edu:11371/pks/lookup?op=get&search=0x81F00126
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <http://lists.nclug.org/pipermail/nclug/attachments/20051015/135f5203/attachment.pgp>