Re: [ojAlgo-user] Heap Space

[Anders P. <an...@op...>]

Are you saying that you’ve successfully executed assignment problems with 8k binary variables using the ojAlgo solvers?

If you want to try and reduce heap space you must first know what it consists of. Run a profiler to learn what is consuming the heap.

Wy guess is that the design of the integer solver and how it interacts with the sub-solvers is not efficient enough for you. (maybe there’s something we can improve, and if you’ve already succeeded with 8k variables...)


/Anders



> On 2 Dec 2016, at 02:26, Karan Singh <kar...@gm...> wrote:
> 
> I apologize for the typo - I meant 8K - 15K variables. 80K would be quite the problem!
> 
> Thanks,
> Karan
> 
> On Thu, Dec 1, 2016 at 5:25 PM, Karan Singh <kar...@gm...> wrote:
> Hello all,
> 
> I'm working on a large (80K - 150K binary variables) general assignment problem with a few constraints & preferences. I am attempting to assign workers to shifts.
> 
> The library has been working well for me, but I recently ran into heap space issue. I found that with 5000 variables, I require 2GB of heap space, and with 8000 variables, I require 4GB of heap space. This problem may scale in the future, so I'd like to limit heap space.
> 
> Does anyone have any suggestions? I'd be happy with a slower solution, if there's a way to perhaps use less space.
> 
> Thanks,
> Karan
> 
> ------------------------------------------------------------------------------
> Check out the vibrant tech community on one of the world's most 
> engaging tech sites, SlashDot.org! http://sdm.link/slashdot_______________________________________________
> ojAlgo-user mailing list
> ojA...@li...
> https://lists.sourceforge.net/lists/listinfo/ojalgo-user