On 4/29/2016 2:33 PM, Bill Bogstad wrote: > make it not work. I suspect that if they hadn't insisted on making > the differences in latency completely invisible to applications, it is > possible that it would have worked better. A way for a program to
Certainly, but but they already had low latency, multi-node memory sharing with Memory Channel. They were trying something new with Galaxy. Sometimes new things don't work out. That's what R&D is all about. -- Rich P. _______________________________________________ Discuss mailing list [email protected] http://lists.blu.org/mailman/listinfo/discuss
