Re: Mathematica 7 is now available
- To: mathgroup at smc.vnet.net
- Subject: [mg94524] Re: Mathematica 7 is now available
- From: Jon Harrop <jon at ffconsultancy.com>
- Date: Mon, 15 Dec 2008 07:47:49 -0500 (EST)
- References: <gg62p3$56g$1@smc.vnet.net> <gh8hkr$r1d$1@smc.vnet.net> <ghavsp$ock$1@smc.vnet.net> <200812081121.GAA15802@smc.vnet.net> <ghlmbp$jof$1@smc.vnet.net> <200812100949.EAA00307@smc.vnet.net> <ghqk4i$193$1@smc.vnet.net> <ghtjjj$red$1@smc.vnet.net> <gi2ucs$9na$1@smc.vnet.net>
Jean-Marc Gulliet wrote:
> Jon Harrop wrote:
>> I've managed to get rid of all the errors by calling
>> DistributeDefinitions with the symbol names of all functions *and* the
>> local "scene" variable right before the ParallelMap. However, this still
>> doesn't give any performance improvement (and it only spawns 4 kernels
>> for my 8 cores).
>
> John,
>
> You seem to believe that Mathematica 7.0 uses as many core as available
> on a given machine (the marketing documentation is quite misleading to
> this respect).
>
> However, the default license allows only up to 4 cores to be used (of
> course, one can upgrade one's license by paying an additional fee).
Yes. I believe I succeeded in parallelizing the ray tracer across 4 of my 8
cores but it gave no speedup. Specifically, I had 1 core at 100% usage,
three more at 20% usage and performance was degraded significantly.
For reference, my final timings of our implementations are:
420,000s My Mathematica
2,500s Daniel's Mathematica
3.9s My serial OCaml
1.4s My parallel OCaml on 8 cores
So the performance difference really was five orders of magnitude as my
preliminary tests had indicated.
--
Dr Jon D Harrop, Flying Frog Consultancy Ltd.
http://www.ffconsultancy.com/?u
- References:
- Re: Mathematica 7 is now available
- From: Jon Harrop <jon@ffconsultancy.com>
- Re: Mathematica 7 is now available
- From: Jon Harrop <jon@ffconsultancy.com>
- Re: Mathematica 7 is now available