[Beowulf] cluster on Mellanox Infiniband (fwd)
Many of your questions may have already been answered in earlier discussions or in the FAQ. The search results page will indicate current discussions as well as past list serves, articles, and papers.
Joachim Worringen joachim at ccrl-nece.deWed Jun 30 01:27:44 PDT 2004
- Previous message: [Beowulf] cluster on Mellanox Infiniband (fwd)
- Next message: [Beowulf] KVM to a compute node - ssh
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Franz Marini: > On Wed, 2004-06-23 at 08:22, Emma wrote: > > --- Franz Marini <franz.marini at mi.infn.it> wrote: > > > We have a small (6 dual Xeon nodes, plus server) > > > testbed cluster with > > > Mellanox Infiniband (switched, obviously). > > > > > > > > > So far, it's been really good. We tested the net > > > performance with SKaMPI4 > > > ( http://liinwww.ira.uka.de/~skampi/ ), the results > > > should be in the > > > online db soon, if you want to check them out. > > > > Can you notice us when the statistics published > > online? > > > > > > Thanks, > > -Bo > > Just a quick note to let you know that the results are online. > The cluster is "Valinor" (yes, Tolkien ;)). Thanks for putting these numbers online. I took a quick glance at the usual values and have to admit that I am not really impressed - a minimal latency of about 15us is not exactly state of the art. However, it seems that Skampi produces "different" numbers than other benchmarks (as usual...). Do you have numbers for mpptest available? And what about running 2 processes per node? Joachim -- Joachim Worringen - NEC C&C research lab St.Augustin fon +49-2241-9252.20 - fax .99 - http://www.ccrl-nece.de
- Previous message: [Beowulf] cluster on Mellanox Infiniband (fwd)
- Next message: [Beowulf] KVM to a compute node - ssh
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Beowulf mailing list
