top500 list (was: opteron VS Itanium 2)

Keith D. Underwood kdunder at sandia.gov
Mon Nov 17 11:56:51 EST 2003


> > Lightning (LANL Opteron)    (2816 CPUs) -  8.1 TFlops
> 
> yield is 72%.

BTW, 60% of 8 GF == 4.8 GF per processor.  72% of 4 GF == 2.88.  If you
use LINPACK as a metric, why do you think the latter wins?  (note the
if. I'm not suggesting LINPACK is the right benchmark.)


> for the billionth time: rmax is just a matter of how much money
> you have.  rmax/rpeak is the only part of top500 that matters.

You have to include cost. Otherwise, if I could buy a 1 PetaFLOP system
that yielded 10% efficiency for the same price as a 100 TeraFLOP system
that yielded 75% efficiency, I should buy the latter?  Or, put another
way, a vendor would be better off building a slower processor with a
modern memory system that achieved 95% of peak.  You can always put more
of them together with more money, right?  (I'm not sure if I know of any
networks that scale to 100,000 processors).  

rmax/rpeak is just as bad (or worse) of a metric as rmax if it is the
only metric.  It's not like LINPACk is terribly communication bound or
anything (in which case, rmax/rpeak might mean something).  

						Keith


_______________________________________________
Beowulf mailing list, Beowulf at beowulf.org
To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf



More information about the Beowulf mailing list