Im sorry for replying so late<div><br></div><div>@Enrique Fynn</div><div><br></div><div>I'm really interesting working with you on that implement.</div><div><br></div><div>GA's could be a master-slave problem master being the evolution program and slave being fitness functions.</div>
<div>I have no knowledge about OpenMPI. I got some great tutorials here: </div><div><a href="https://rqchp.ca/?mod=cms&lang=EN&navrev=off&version=-1&pageId=96&">https://rqchp.ca/?mod=cms&lang=EN&navrev=off&version=-1&pageId=96&</a></div>
<div><a href="https://rqchp.ca/modules/cms/checkFileAccess.php?file=local.rqchpweb_udes/mpi/exemples_cpp/ex16_request_EN.cpp">https://rqchp.ca/modules/cms/checkFileAccess.php?file=local.rqchpweb_udes/mpi/exemples_cpp/ex16_request_EN.cpp</a> </div>
<div><br></div><div>P.S. I'm in Campinas Brasil</div><div><br></div><div>@ Blazefirst</div><div><br></div><div>thanks for your reply. I have a single objective function with double precision. I saw that double precision was only supported by one type of nvidia card :-(.</div>
<div><br></div><div>I saw your paper about weekly menu planning and that's a "really good piece of meat !"<br><br><div class="gmail_quote">2010/5/4 <span dir="ltr"><<a href="mailto:galib-request@mit.edu">galib-request@mit.edu</a>></span><br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">Send galib mailing list submissions to<br>
<a href="mailto:galib@mit.edu">galib@mit.edu</a><br>
<br>
To subscribe or unsubscribe via the World Wide Web, visit<br>
<a href="http://mailman.mit.edu/mailman/listinfo/galib" target="_blank">http://mailman.mit.edu/mailman/listinfo/galib</a><br>
or, via email, send a message with subject or body 'help' to<br>
<a href="mailto:galib-request@mit.edu">galib-request@mit.edu</a><br>
<br>
You can reach the person managing the list at<br>
<a href="mailto:galib-owner@mit.edu">galib-owner@mit.edu</a><br>
<br>
When replying, please edit your Subject line so it is more specific<br>
than "Re: Contents of galib digest..."<br>
<br>
<br>
Today's Topics:<br>
<br>
1. GaLib - Multithreading (Guillaume Mass?)<br>
2. Re: GaLib - Multithreading (Enrique Fynn)<br>
3. Re: GaLib - Multithreading (Blazefirst)<br>
<br>
<br>
----------------------------------------------------------------------<br>
<br>
Message: 1<br>
Date: Mon, 3 May 2010 19:18:51 -0300<br>
From: Guillaume Mass? <<a href="mailto:guillaume.masse@polymtl.ca">guillaume.masse@polymtl.ca</a>><br>
Subject: [galib] GaLib - Multithreading<br>
To: <a href="mailto:galib@mit.edu">galib@mit.edu</a><br>
Message-ID:<br>
<<a href="mailto:h2y8229c9771005031518q20deb75bw3a2707e68aebcc61@mail.gmail.com">h2y8229c9771005031518q20deb75bw3a2707e68aebcc61@mail.gmail.com</a>><br>
Content-Type: text/plain; charset="iso-8859-1"<br>
<br>
Hi,<br>
<br>
I Have a expensive calculation of the objective function in my work.<br>
I want to know if there is anything in the lib that can do parallel<br>
computation of populations.<br>
Is there anyone working on this feature?<br>
Is there anyone who would like to work on that with me ?<br>
<br>
Guillaume Mass?<br>
?cole polytechnique de Montr?al<br>
-------------- next part --------------<br>
An HTML attachment was scrubbed...<br>
URL: <a href="http://mailman.mit.edu/pipermail/galib/attachments/20100503/bfd5c86c/attachment-0001.htm" target="_blank">http://mailman.mit.edu/pipermail/galib/attachments/20100503/bfd5c86c/attachment-0001.htm</a><br>
<br>
------------------------------<br>
<br>
Message: 2<br>
Date: Mon, 3 May 2010 19:24:52 -0300<br>
From: Enrique Fynn <<a href="mailto:enriquefynn@gmail.com">enriquefynn@gmail.com</a>><br>
Subject: Re: [galib] GaLib - Multithreading<br>
To: Guillaume Mass? <<a href="mailto:guillaume.masse@polymtl.ca">guillaume.masse@polymtl.ca</a>><br>
Cc: <a href="mailto:galib@mit.edu">galib@mit.edu</a><br>
Message-ID:<br>
<<a href="mailto:k2g97eb84651005031524x9d741ffcpfde5aa4792559263@mail.gmail.com">k2g97eb84651005031524x9d741ffcpfde5aa4792559263@mail.gmail.com</a>><br>
Content-Type: text/plain; charset="utf-8"<br>
<br>
I'm thinking in doing a ga lib. using OpenMPI.<br>
<br>
<br>
Att;<br>
Enrique Fynn.<br>
--<br>
Artificial intelligence laboratory<br>
Federal University of Uberl?ndia<br>
"Society, you're a crazy breed... Hope you're not lonely without me..."<br>
<br>
<br>
<br>
2010/5/3 Guillaume Mass? <<a href="mailto:guillaume.masse@polymtl.ca">guillaume.masse@polymtl.ca</a>><br>
<br>
> Hi,<br>
><br>
> I Have a expensive calculation of the objective function in my work.<br>
> I want to know if there is anything in the lib that can do parallel<br>
> computation of populations.<br>
> Is there anyone working on this feature?<br>
> Is there anyone who would like to work on that with me ?<br>
><br>
> Guillaume Mass?<br>
> ?cole polytechnique de Montr?al<br>
><br>
> _______________________________________________<br>
> galib mailing list<br>
> <a href="mailto:galib@mit.edu">galib@mit.edu</a><br>
> <a href="http://mailman.mit.edu/mailman/listinfo/galib" target="_blank">http://mailman.mit.edu/mailman/listinfo/galib</a><br>
><br>
><br>
-------------- next part --------------<br>
An HTML attachment was scrubbed...<br>
URL: <a href="http://mailman.mit.edu/pipermail/galib/attachments/20100503/b23a5027/attachment-0001.htm" target="_blank">http://mailman.mit.edu/pipermail/galib/attachments/20100503/b23a5027/attachment-0001.htm</a><br>
<br>
------------------------------<br>
<br>
Message: 3<br>
Date: Tue, 04 May 2010 10:43:19 +0200<br>
From: Blazefirst <<a href="mailto:blazefirst@freemail.hu">blazefirst@freemail.hu</a>><br>
Subject: Re: [galib] GaLib - Multithreading<br>
To: <a href="mailto:galib@mit.edu">galib@mit.edu</a><br>
Message-ID: <<a href="mailto:4BDFDE27.7070503@freemail.hu">4BDFDE27.7070503@freemail.hu</a>><br>
Content-Type: text/plain; charset="iso-8859-1"<br>
<br>
Hi,<br>
<br>
we've just implemented and tested a CUDA based parallelization of<br>
fitness calculations<br>
(and also the calculation of the pareto fronts in case of<br>
multi-objective GAs - see <a href="http://mogalib.uni-pannon.hu" target="_blank">http://mogalib.uni-pannon.hu</a>).<br>
<br>
Parallelization of fitness calculation with a GPU might be the way to go<br>
for you, in case your fitness function is<br>
massively parallelizable. If it is, you have to implement the fitness<br>
function in CUDA, and I can provide<br>
you with our modified GALib version.<br>
<br>
Kind regards,<br>
Balazs<br>
<br>
<br>
2010.05.04. 0:18 keltez?ssel, Guillaume Mass? ?rta:<br>
> Hi,<br>
><br>
> I Have a expensive calculation of the objective function in my work.<br>
> I want to know if there is anything in the lib that can do parallel<br>
> computation of populations.<br>
> Is there anyone working on this feature?<br>
> Is there anyone who would like to work on that with me ?<br>
><br>
> Guillaume Mass?<br>
> ?cole polytechnique de Montr?al<br>
><br>
><br>
> _______________________________________________<br>
> galib mailing list<br>
> <a href="mailto:galib@mit.edu">galib@mit.edu</a><br>
> <a href="http://mailman.mit.edu/mailman/listinfo/galib" target="_blank">http://mailman.mit.edu/mailman/listinfo/galib</a><br>
><br>
<br>
-------------- next part --------------<br>
An HTML attachment was scrubbed...<br>
URL: <a href="http://mailman.mit.edu/pipermail/galib/attachments/20100504/96c1a1a0/attachment-0001.htm" target="_blank">http://mailman.mit.edu/pipermail/galib/attachments/20100504/96c1a1a0/attachment-0001.htm</a><br>
<br>
------------------------------<br>
<br>
_______________________________________________<br>
galib mailing list<br>
<a href="mailto:galib@mit.edu">galib@mit.edu</a><br>
<a href="http://mailman.mit.edu/mailman/listinfo/galib" target="_blank">http://mailman.mit.edu/mailman/listinfo/galib</a><br>
<br>
<br>
End of galib Digest, Vol 80, Issue 1<br>
************************************<br>
</blockquote></div><br></div>