Has anyone done a large (2000+) 'beowulf' style compute
cluster with NetBSD? I've been reading a about
some of the expierences from doing large clusters at
Sandia [1].

There are a number of interesting tweek points for 
doing this like; the way paging is done, network
buffering, and a change in the way they do system
calls that are optimized for a net-centric environment.

Do you think that NetBSD would be good for this?

[1] http://www.supercomp.org/sc97/proceedings/TECH/GREENBER/INDEX.HTM

Nate Gelbard		      	       cell: 503.805.1015
Straylight Research		       fax : 516.828.8533

#!/bin/perl -sp0777i<X+d*lMLa^*lN%0]dsXx++lMlN/dsM0<j]dsj
$/=unpack('H*',$_);$_=`echo 16dio\U$k"SK$/SM$n\EsN0p[lN*1
lK[d2%Sa2/d0$^Ixp"|dc`;s/\W//g;$_=pack('H*',/((..)*)$/)