[MITgcm-support] beowulf clusters
chris hill
cnh at mit.edu
Tue Sep 14 10:46:44 EDT 2004
Hi All,
Does anybody have access to one of these? I believe they use the
Transmeta processors which is why you can get a lot of them in a compact
box. I don't know what performance is like though.
Chris
On Tue, 2004-09-14 at 10:04, samar khatiwala wrote:
> Just my $.02 worth:
>
> There was a piece in the NY Times a week or so ago about a company selling
> a cluster in a small box. The really new thing about it was the relatively
> miniscule power consumption (220W for a 12 node cluster). It sounded
> intriguing.
>
> http://www.orionmulti.com
>
> Samar
>
> On Tue, 14 Sep 2004, Chris Hill wrote:
>
> > Hi Mike,
> >
> > At the moment, the most popular thing, among our vast list of users :-),
> > seems to be Opteron based systems. The Opteron duals tend to do better
> > scaling than the Xeon duals. There are also some G5 cluster advocates out
> > there too. They may want to comment?
> >
> > Chris
> >
> > > -----Original Message-----
> > > From: mitgcm-support-bounces at mitgcm.org
> > > [mailto:mitgcm-support-bounces at mitgcm.org] On Behalf Of Mike Spall
> > > Sent: Tuesday, September 14, 2004 9:47 AM
> > > To: mitgcm-support at mitgcm.org
> > > Subject: [MITgcm-support] beowulf clusters
> > >
> > > I am in the process of shopping for a small (4 compute nodes,
> > > 8 processors) beowulf cluster. I have been getting quotes
> > > from PSSC Labs, Aspen Systems, and Microway. I was wondering
> > > if anyone has done any benchmarking of the MITgcm with
> > > similar machines.
> > > Any preference between Xeon and Opteron? I am looking at the
> > > GB ethernet connection. The prices I am getting are 10-15k.
> > > Any suggestions would be welcome.
> > >
> > > Thanks,
> > > Mike
> > >
> > > --
> > > ==================================================
> > >
> > > Michael A. Spall
> > > Department of Physical Oceanography
> > > W.H.O.I. MS #21
> > > 360 Woods Hole Road
> > > Woods Hole, MA 02543
> > >
> > > mspall at whoi.edu
> > > (508) 289-3342 (office)
> > > (508) 457-2181 (fax)
> > >
> > > =================================================
> > >
> > > _______________________________________________
> > > MITgcm-support mailing list
> > > MITgcm-support at mitgcm.org
> > > http://dev.mitgcm.org/mailman/listinfo/mitgcm-support
> > >
> >
> > _______________________________________________
> > MITgcm-support mailing list
> > MITgcm-support at mitgcm.org
> > http://dev.mitgcm.org/mailman/listinfo/mitgcm-support
> >
> _______________________________________________
> MITgcm-support mailing list
> MITgcm-support at mitgcm.org
> http://dev.mitgcm.org/mailman/listinfo/mitgcm-support
More information about the MITgcm-support
mailing list