[SunRescue] Re: New Netra

[Cyrus M. Reed]

Along with the other uses mentioned, they also work well for Quantum
Mechanics problems.  Somehow the parallel nature of the architecture works
really well on the sparce matrices encountered in such problems (or so
I've been told).  I know someone at my university that is prototyping a
Beowulf cluster with some spare machines for just such a use.  If the
prototype works (sort of a "proof of concept"), then some real money will
probably be put into a full setup.  I think the majority of the "apps" you
will find for Beowulf clusters have been home grown by someone who wants a
particular problem solved.

-Cyrus

On Wed, 17 Jan 2001, Reagen Ward wrote:

> On Wed, Jan 17, 2001 at 07:52:45PM -0800, Cyrus M. Reed wrote:
> > in there.  That wouldn't be so bad, considering the overall price.  You
> > could make a cool Beowulf cluster out of them too...  (ducks rotten
> > fruit...)
> 
> I know you were kidding, but it made me think of something.  I regularly 
> hear comments about building Beowulf clusters, and maybe one of these days
> I'll get someone to tell me just what they would do with a one.  So few 
> apps can use MPI / PVM / whatever, the uses of one are quite limited.
> 
> Reagen
> _______________________________________________
> Rescue maillist  -  Rescue at sunhelp.org
> http://www.sunhelp.org/mailman/listinfo/rescue
>