Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: crafty speedup numbers

Author: Robert Hyatt

Date: 17:02:50 05/06/04

Go up one level in this thread


On May 06, 2004 at 19:03:48, martin fierz wrote:

>aloha!
>
>bob posted some crafty logfiles running a 24-position test set on his ftp site
>(for anyone else crazy enough to repeat what i did:
>ftp.cis.uab.edu/pub/hyatt/smpdata)
>
>these are logfiles of crafty running as single CPU, dual, or quad; on opterons.
>i took the last completed ply on the single CPU set for each position (marked by
>-> in the logfile, i hope...), wrote down the time to complete this ply, and did
>this for all logfiles. there are 9 of these, 4 repeats for 2 and 4 CPUs. i
>computed the speedup for time-to-finish-ply-X for each of the multi-CPU runs
>with the following results:
>
>2 CPUs:
>1.961 +- 0.093
>1.888 +- 0.074
>1.846 +- 0.078
>1.763 +- 0.084
>
>4 CPUs:
>3.15 +- 0.15
>3.29 +- 0.20
>3.06 +- 0.12
>3.19 +- 0.13
>

That is higher than my number although I only checked log 1 for mt=4.  Did you
compute the speedup for each position, then add and divide by 24?  If so, I'm
not a fan of that way.  A long search on an efficient position skews the
results.  I prefer to take the total time for each run and use that...

But either is "a number"...


>now, is there any meaning to this, and if yes, what?
>
>point #1 to make is that the numbers here are mutually consistent with each
>other, given the error margins quoted. which should show those skeptical of this
>statistical approach that it makes sense to do it this way, rather than to just
>write "i measured speedup 3.1".
>
>point #2 is that the speedup on 4 CPUs on average is 3.17 in this test, which
>might be one point for bob in the duel with vincent; although i suspect that the
>speedup depends on the hardware architecture - i will leave this question to the
>parallel computing experts though...

The opteron is not as good as a real SMP box.  Such as my quad xeons.  Opteron
is NUMA which definitely has "issues" to deal with.


>
>point #3 is perhaps most important for the bob vs vincent duel: the standard
>error for a 4 CPU test run is on the order of 0.2. if vincent's tests were with
>a similarly small number of positions, then the differences measured in these
>experiments (2.8 / 3.0 / 3.1) are statistically insignificant, and the whole
>argument is pointless :-)



I have said that 100 times.  But let's see if you convince Vincent...




>
>cheers
>  martin
>
>
>disclaimer: i computed the search time in seconds from crafty's log file by
>converting minutes:seconds to seconds in my head. i may have made a mistake here
>or there, although i did my best not to - but it's late at night and quite
>boring to look through crafty logfiles...



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.