RE: [Help-glpk] General, parallel Dantzig-Wolfe implementation now available

[Joey Rios]

Hi Robbie,

>
> I added some details about your Dantzig-Wolfe project
> to the GLPK wikibook:
>
> http://en.wikibooks.org/wiki/GLPK/Add-Ons#Dantzig-Wolfe_decomposition
>

Great!  Thanks for adding it.  Write-up looks good to me.

> In passing, DW was suggested for a project I used to
> work on. Indeed the suggestion was published in
> 'Energy', 20(9):941-958, in 1995, but never implemented.
> Not uncommon in academia I must add.

This is exactly the motivating factor in releasing the code.  If researchers 
don't have to go through the work of implementing the algorithm, I think it 
might get even more use across research domains.

On a related note, I know a few people have downloaded the code already.  I'd 
be happy to hear if they actually got it to build and could at least run the 
examples.  After that, if anyone wants to let me know what they think of the 
utility of the implementation or places for near-term improvement, I'd love to 
hear about it.

>
> > Hi all,
> >
> > I had a private email exchange with Andrew a little
> > over a year ago regarding a Dantzig-Wolfe
> > implementation built upon GLPK. Well, the legal
> > department at work has finally allowed me to release
> > it. It can be found currently on SourceForge:
> >
> > http://sourceforge.net/projects/dwsolver/


Joey