Results included in attached file...
Mike.
> (3) I would be interested in seeing the STREAM performance numbers for the Xeon Phi 7120P. We have one "revision C" Xeon Phi at TACC, but its STREAM numbers are a bit lower than the values I get on the Xeon Phi SE10P. I seem to recall that Intel knew about this, but I don't remember if they provided an explanation. The Xeon Phi 7120P should probably use the same optimizations as the published Xeon Phi SE10P result:
Received on Thu Jun 04 2015 - 13:13:15 CDT