[Beowulf] /. Cooler room or cooler servers?

Mark Hahn hahn at physics.mcmaster.ca
Thu Apr 7 17:14:37 PDT 2005


> > > One also needs to look at important ancillary issues like power 
> > consumption
> > > of the cache, bus drivers, etc.  And, of course, instruction stream 
> > makes a
> > > big difference.
> >
> >kind of obvious, don't you think?  when I said peak, I meant peak,
> >and I'm quoting power dissipation measured at the wall plug (kill-a-watt).
> >for these systems, running two compute-intensive programs maxes out;
> >messing with memory or disks only adds a few watts (percent).
> 
> My point was that the "Superior Power Efficiency" kind of chart (no matter 
> where the numbers come from) is really meaningless.  Simply knowing the 
> basic technology level will tell you how many Joules/calculation you'll get 
> to a first order.  Any finer refinement will be very, very implementation 
> and algorithm dependent.

I guess it's hard to argue with a "to a first order" statement.
but right now, AMD and Intel high-end chips are pretty comparable 
in almost any performance metric, but AMD is noticably cooler.

this is not meaningless.




More information about the Beowulf mailing list