[openib-general] ip over ib throughtput

[Ronald G. Minnich]

would you all recommend just cutting over to openib at this point? I've
got a 256-node cluster with hierarchical switches that has never really
worked well with the vendor stacks, and the opensm has never run more than
one day without segv; is openib solid enough that I can run with it? Is
the opensm in openib pretty solid? Are there tools for tracing the fabric 
and looking for problems?

thanks

ron