Re: [ojAlgo-user] How to set OjAlgoUtils.ENVIRONMENT?

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

I’ve made sure that future versions of ojAlgo will recognise a hardware configuration like that.

Regarding your LP taking too long - this is not the solution to that.

The LP solver needs to be rewritten using sparse data structures. That's normally what becomes the bottleneck when scaling up the problem size, but I would think 500 variables should have been ok.

When ojAlgo’s built in solvers are not enough you can switch to other solvers without changing any code. I believe the community edition of CPLEX allows up to 1000 variables. If you want to try that just follow there instructions:

https://github.com/optimatika/ojAlgo-extensions/tree/master/ojAlgo-cplex


/Anders



> On 3 Jul 2017, at 14:52, Leon Graser via ojAlgo-user <oja...@li...> wrote:
> 
> Hey Anders,
> 
> my first approach was to use the "Hardware.makeSimple()" call mentioned below. But it did not work the way I expected it to do 😊
> 
> I'm running my optimization on an Intel i5-4670K with 16GB of RAM which I like to make use of.
> 
> Best regards,
> Leon
> 
> -----Ursprüngliche Nachricht-----
> Von: Anders Peterson [mailto:an...@op...] 
> Gesendet: Samstag, 1. Juli 2017 22:29
> An: Leon Graser <st1...@st...>
> Cc: oja...@li...
> Betreff: Re: [ojAlgo-user] How to set OjAlgoUtils.ENVIRONMENT?
> 
> What you see is not an error - it’s just info.
> 
> You get that info whenever you “start” ojAlgo and its predefined hardware profiles don’t match. If you reset OjAlgoUtils.ENVIRONMENT that happens after, and there is no need to set it if all you do is a “simple”.
> 
> Take a look at the source code of OjAlgoUtils and you should see how it works:
> 
> https://github.com/optimatika/ojAlgo/blob/master/src/org/ojalgo/OjAlgoUtils.java
> 
> 
> What you could do is tell me what kind of cpu and how much ram you have, and I’ll make sure ojAlgo understands that.
> 
> 
> /Anders
> 
> 
>> On 1 Jul 2017, at 20:58, Leon Graser via ojAlgo-user <oja...@li...> wrote:
>> 
>> Hey,
>> 
>> I managed to solve small linear optimization problems using ojAlgo. Unfortunately, for bigger problems of approx. 450 variables and up it takes ages and the memory limits the execution. Hence I increased the memory using the -Xmx JVM argument. But now I get an error stating to modify “OjAlgoUtils.ENVIRONMENT” according to my system. I tried to use “Hardware.makeSimple("x86_64", 17179869184l, 4).virtualise();” but I still receive the very same error. 
>> Any hints how to set the environment field correctly?
>> 
>> Best regards,
>> Leon
>> ----------------------------------------------------------------------
>> -------- 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
> 
> 
> 
> ------------------------------------------------------------------------------
> 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