[Beowulf] massive parallel processing application required

Robert G. Brown rgb at phy.duke.edu
Wed Jan 31 07:55:31 PST 2007

On Wed, 31 Jan 2007, Mark Hahn wrote:

>> Climate code, especially when coupling between atmosphere and ocean models?
> seems like it would require some nontrivial physics, not to mention
> realistic input data.  don't most climate codes also depend on huge
> multi-dimensional FFT's where the transpose is coded as all-to-all?
> here's an alternative: nbody physics.  just put a bunch of particles in some
> empty space and see what they do as they interact through gravity.  of
> course, gravity is all-to-all, but then again in a nontrivial sense, less 
> coupled problems are less interesting...

Or another simple physics problem -- simulate e.g. the Ising problem, or
any of a number of problems in magnetism.  Nearest neighbor
interactions, "known results".

And if it is just a matter of a nifty demo, don't forget the always
useful parallel mandelbrot set packages and/or rendering packages
(povray).  I'm pretty sure both are still around -- I still use xep
to demo PVM, although I have hacked it a bit because it is now too easy
to get to the "bottom" of floating point resolution even on a single
processor, which actually kind of sucks as the display breaks down just
when you get way down into the set where things get very odd and spiky.
Spikier.  Whatever.


> _______________________________________________
> Beowulf mailing list, Beowulf at beowulf.org
> To change your subscription (digest mode or unsubscribe) visit 
> http://www.beowulf.org/mailman/listinfo/beowulf

Robert G. Brown	                       http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567  Fax: 919-660-2525     email:rgb at phy.duke.edu

More information about the Beowulf mailing list