[Cbc] Re : Parallel version of Cbc

Joel Joel joel_31415 at yahoo.fr
Mon Oct 6 12:29:49 EDT 2008


Thanks for your answer, Stefan. 

I have commented the line 112 of the file CbcModel.cpp in order to undefine "small" as you suggested and it compiled fine.

When I look at the CPU activity on my dual-core during the branch-and-cut, apparently both CPU seems to work, but the total CPU usage remains close to 50%.
The bevahior seems very similar to the non-parallel version of Cbc... 

Do I have to change something when I call Cbc from my code (to indicate to Cbc, for example, the number of CPU) ? 

Thanks again!

Joel.





----- Message d'origine ----
De : Stefan Vigerske <stefan at math.hu-berlin.de>
À : Joel Joel <joel_31415 at yahoo.fr>
Cc : cbc at list.coin-or.org
Envoyé le : Lundi, 6 Octobre 2008, 17h23mn 28s
Objet : Re: [Cbc] Parallel version of Cbc

Hi,

> I have compiled Cbc (SVN revision 1087) with MinGW, using the the "standard" procedure: configure, make, make 
install.
> I am able to use Cbc as a static library and it works fine. 
> 
> I wanted to try the parallel version of Cbc, using: ".\configure --enable-cbc-parallel".
> 
> 1°) During the configuration script, I have the following warning message: 
>     "-lrt unavailable; using gettimeofday instead of clock_gettime"
> 
> 2°) During Cbc compilation, I have an error message:
> [....]
> CbcModel.cpp:7869: error: expected unqualified-id before '=' token

bool small = size<=550;

This more looks like a problem with including windows.h (which, as Laci 
found out, defines "small"). Maybe it get included by pthread.h ?

Therefore, try a
#undef small
somewhere before.

> Could anyone give me some help?
> - Is the librt library mandatory?  If yes, where can I download it? 

It is used for time measuring.
For Cbc/stable/2.2, librt is optional, but I don't know if the 
corresponding changes have made it into Cbc/trunk.

> - Is it currently possible to compile a parallel version of Cbc or is it too early? If yes, what am I doing wrong? 

Works fine for me.

Stefan

-- 
Stefan Vigerske
Humboldt University Berlin, Numerical Mathematics
http://www.math.hu-berlin.de/~stefan



      
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://list.coin-or.org/pipermail/cbc/attachments/20081006/1a114f65/attachment.html 


More information about the Cbc mailing list