Author: Robert Hyatt
Date: 20:24:04 07/19/00
Go up one level in this thread
On July 19, 2000 at 13:05:41, José de Jesús García Ruvalcaba wrote: >On July 18, 2000 at 19:51:47, Chris Carson wrote: > >>I wonder how much better it would do? >> >>8x 700 = 5600 >>32x 300 = 9600 >> >>Just a hypothetical. :) >> >>I guess 50 points. Any other known Intel >>configurations that might yeild better >>results. Intel anounced a 1000 processor >>configuration at one time (super computer). >>I would love to see DJ 6 on that system. >>I this is at Los Alamos NM. :) >> >>Best Regards, >>Chris Carson > > P.ConNerS hardware is not SMP. Any parallel search on it will be less efficient >that Deep Junior's parallel search, which is SMP (not to mention the required >major rewrite). > My guess is that if Deep Junior is rewritten to run on P.ConNerS hardware, it >will be just as good as in the eight-way Xeon due to the unavoidable search >inefficiencies. > I would be more interested to see P.ConNerS running on a Siemens Primergy, of >course after it is redesigned to take advantage of it. >José. It should run on a primergy, or dell, or any 8-way box with no changes. Message-passing works fine there as there is no network latency at all. Of course, a shared memory implementation would be even faster, but that is the nice thing about cluster applications... they run seamlessly on SMP hardware, while vice-versa doesn't happen easily.
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.