On 3 March 2011 21:56, Calcpage <calcp...@aol.com> wrote: > Well, I'd like to expose my Computing Independent Study (aka CIStheta) class > to MPI. I was wondering if it is easier to set up under SAGE. > I have run live Linux CD solutions for clustering for years now such as > ClusterKnoppix, BCCD and Quantian for openMosix and PVM. That was years ago. > More recently we've used parallelKnoppix, pelicanHPC and ClusterByNight for > MPI. I was thinking of setting up something more permanent, however. > BTW, our adhoc cluster has 25 64bit dualcore AMD Athlons running 64bit > Ubuntu Desktop over a gigE switch. I did a little benchmarking recently > using pelicanHPC and ran 50 cores at about 24 GFLOPS. That's not fast for a > cluster these days, but fine for educational demonstrations I think. Let's > face it, IBM's Watson runs at 72 TFLOPS which 3000 times faster, but my > cluster doesn't have to play Jeopardy! > Regards, > A. Jorge Garcia
As I remarked on a comment on http://trac.sagemath.org/sage_trac/ticket/10869 I feel the OpenMPI optional package needs to be documented better. OpenMPI is an optional package in Sage, but there's no real clues how to use it. No examples. Now compare that to the documentation for Mathematica on parallel computing http://reference.wolfram.com/mathematica/ParallelTools/tutorial/Overview.html We have a library which should allow parallel computations, but there's nothing to help a user achieve this. Dave -- To post to this group, send email to sage-support@googlegroups.com To unsubscribe from this group, send email to sage-support+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-support URL: http://www.sagemath.org
