Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: parallel scaling

Author: Robert Hyatt

Date: 06:51:27 10/30/03

Go up one level in this thread


On October 30, 2003 at 08:14:24, Vincent Diepeveen wrote:

>On October 30, 2003 at 00:44:54, Dave Gomboc wrote:
>
>>On October 29, 2003 at 17:37:08, Robert Hyatt wrote:
>>
>>>On October 29, 2003 at 14:20:01, Vincent Diepeveen wrote:
>>>
>>>>On October 28, 2003 at 23:21:55, Robert Hyatt wrote:
>>>>
>>>>>On October 28, 2003 at 18:12:16, Vincent Diepeveen wrote:
>>>>>
>>>>>>On October 28, 2003 at 09:48:52, Robert Hyatt wrote:
>>>>>>
>>>>>>>On October 27, 2003 at 21:23:13, Vincent Diepeveen wrote:
>>>>>>>
>>>>>>>>On October 27, 2003 at 20:09:55, Eugene Nalimov wrote:
>>>>>>>>
>>>>>>>>>On October 27, 2003 at 20:00:54, Robert Hyatt wrote:
>>>>>>>>>
>>>>>>>>>>On October 27, 2003 at 19:57:12, Eugene Nalimov wrote:
>>>>>>>>>>
>>>>>>>>>>>On October 27, 2003 at 19:24:10, Peter Skinner wrote:
>>>>>>>>>>>
>>>>>>>>>>>>On October 27, 2003 at 19:06:51, Eugene Nalimov wrote:
>>>>>>>>>>>>
>>>>>>>>>>>>>
>>>>>>>>>>>>>I don't think you should be afraid. 500 CPUs is not enough -- you need
>>>>>>>>>>>>>reasonable good program to run on them.
>>>>>>>>>>>>>
>>>>>>>>>>>>>Thanks,
>>>>>>>>>>>>>Eugene
>>>>>>>>>>>>
>>>>>>>>>>>>I would bet on Crafty with 500 processors. That is for sure. I know it is quite
>>>>>>>>>>>>a capable program :)
>>>>>>>>>>>>
>>>>>>>>>>>>Peter.
>>>>>>>>>>>
>>>>>>>>>>>Efficiently utilizing 500 CPUs is *very* non-trivial task. I believe Bob can do
>>>>>>>>>>>it, but it will be nor quick nor easy.
>>>>>>>>>>>
>>>>>>>>>>>Thanks,
>>>>>>>>>>>Eugene
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>If the NUMA stuff doesn't swamp me.  And if your continual updates to the
>>>>>>>>>>endgame tables doesn't swamp me.  We _might_ see some progress here.  :)
>>>>>>>>>>
>>>>>>>>>>If I can just figure out how to malloc() the hash tables reasonably on your
>>>>>>>>>>NUMA platform, without wrecking everything, that will be a step...
>>>>>>>>>
>>>>>>>>>Ok, just call the memory allocation function exactly where you are calling it
>>>>>>>>>now, and then let the user issue "mt" command before "hash" and "hashp" if (s)he
>>>>>>>>>want good scaling.
>>>>>>>>>
>>>>>>>>>Thanks,
>>>>>>>>>Eugene
>>>>>>>>
>>>>>>>>That's why i'm multiprocessing. All problems solved at once :)
>>>>>>>
>>>>>>>
>>>>>>>And several added.  Duplicate code.  Duplicate LRU egtb buffers.  Threads
>>>>>>
>>>>>>Duplicate code is good. Duplicate indexation egtb tables is good too (note the
>>>>>>DIEP ones do not require 200MB for 6 men, but a few hundreds of KB only).
>>>>>>
>>>>>
>>>>>wanna compare access speeds for decompression on the fly?  If you make
>>>>>the indices smaller, you take a big speed hit.  It is a trade-off.
>>>>
>>>>Not really, I need compressed around 500MB for all 5 men. Nalimov 7.5GB.
>>>>
>>>>What's more compact?
>>>
>>>Let's compare apples to apples.  You are storing DTM for all 3-4-5 piece
>>>files in 500MB?  You just set a new world record for size.
>>>
>>>Aha.  You aren't storing DTM, you are storing W/L/draw?  Then the comparison
>>>is not equal.
>>>
>>>Either way, my statement stands...
>>
>>It sounds to me like he's storing enough that he can do DTM with some lookahead
>>search.
>>
>>Dave
>
>Add a note to all this. In 1999 EGTBs would have helped diep really a lot.
>In 2002 WM they only hurted performance only. Endgames of engines is well enough
>now to get into positions where EGTBs can potentially nail them.
>
>Of course it still happens but the significance of EGTBs for world champs 2003
>and further will not be too big.
>
>Nevertheless i am sure Bob is going to deny this, pointing to some toy games at
>5 0 at ICC.
>
>Note that Bob is incorrect in his assumption that 7.5GB is better than 500MB and
>that against 200MB for indice is better than a few hundred kilobytes.
>
>I always wondered for a year or 10 why bob is doing such assumptions here. I see
>it as a classical Hyatt-debating technique.
>
>"let your opponent proof everything from ground up".
>


It is just called "not believing hand-waving and promises from someone who
has _never_ delivered anything in the past.

hehehehe...




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.