[Beowulf] Nehalem Xeons 55** series

[Mark Hahn]

> AMD Opteron "Shangai" 6MB L3 : 90% of peak
> Intel "Nehalem" Core i7 : 95.2% of peak
> Intel Itanium 2 : 95.6% of peak

interesting numbers, and Goto's efforts are always respected.
it would be valuable to understand why, though.  I usually think
of ia64 as being basically designed specifically to make this kind 
of hero-optimization possible.  but I wonder how well each of these 
chips does on low-to-medium quality user code and current compilers.

> It's a Core i7 but should give you an idea.

I suspect that lots of people are holding their breaths (and POs)
waiting for Intel to release 2 (and 4?) socket nehalems.  it has to be 
a bit scary for AMD, since they're getting leapfrogged in the area
of their traditional strength.  I don't have a sense for whether Goto's
kind of tuning (beyond the ability of compilers, and relatively
cache-friendly relative to flops) is terribly relevant to the market.

on the topic of memory bandwidth:
  http://techreport.com/articles.x/16448
makes it sound as if AMD knows they have a scaling problem with cache
coherency, and have a solution queued.  does anyone know whether nehalem
already has a probe filter?  or if AMD has mentioned anything about 
widening to a 3-4-dimm-per-socket memory interface?

regards, mark hahn.