Souper Computer

[jrv at vanzandt.mv.com (James R. Van Zandt)]

Scott Lipcon <slipcon at cs.jhu.edu> writes:

>Our network is 
>100Mb switched ethernet.  I've heard of people using normal 10Mb 
>ethernet for smaller clusters, that they're setting up just for fun, 
>but it probably doesn't work too great.

It depends on what problem you attack.  Two I would suggest are:

 - Monte Carlo simulations, where you need 100s or 1000s of iterations,
differing only in the random number seed.

 - Genetic algorithms, where 10e3 to 10e6 population members have to
be evaluated over the course of the run.

Assuming each of the calculations is substantial (minutes?),
communications should not dominate.

			  - Jim Van Zandt
-
Subcription/unsubscription/info requests: send e-mail with subject of
"subscribe", "unsubscribe", or "info" to discuss-request at blu.org