> The cluster was build of a various types of machines,
> including Alphas, Xeons, and "conventinals" ..
The conventional probably were the workhorses even though not fastest
per CPU.
> As "workload" they rendered 2 MPG3 movies, executed
> GFLOPS benchmarks, solved 80th-degree polynoms and
> run various other programs.
It was a 8000th degree polynom ...
The benchmark was Linpack and we did run it only on a subcluster which
consisted of 50 666MHz 21164 Alphas with 256mb RAM each. So while
the result is respactable fast (we for shure shreddered the dusty Cray 2
standing nearby in the exhibition ...) it's by no way a new Linux record
for Linpack. Nevertheless the CPU power made raytracing fun.
I know people have a lot of questions about this event; we intend to
post a writeup with all the details once we all got back the sleep we
didn't get the last days.
Once again thanks to everybody who was helping to make this possible;
some of the people did way more than one can expect somebody to do.
Ralf
-- Linux - where no man has computed before- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/