[Beowulf] New HPCC results, and an MX question

[Joachim Worringen]

Vincent Diepeveen wrote:
> Additional there will be software layers that have to lock in some way.

I'd say all relevant contemporary high-performance interconnects don't 
do any locking for MPI communication. What are you refering to exactly?

  Joachim

-- 
Joachim Worringen - NEC C&C research lab St.Augustin
fon +49-2241-9252.20 - fax .99 - http://www.ccrl-nece.de