[R-sig-eco] large-scale individual-based simulation

[Julian M. Burgos]

MASON is another good option.  It requires some knowledge of Java, though.

http://cs.gmu.edu/~eclab/projects/mason/

David Valentim Dias writes:

> David,
>
> I was stuck with a similar problem. I tried creating LUCC models and R
> can't do the job in the required speed and or just stop working because of
> lack of memory.
> Problem with R is default type is double even for integers and this
> behaviour will eat all memory very fast for 10⁶ agents, you need to think
> about space where they "move/action" too. If you can afford huge memory
> speed can be problematic too.
> I switched to Netlogo for this type of model and the results are surprising
> good. Memory can be a problem in Netlogo too but speed isn't.
> There is another ABM softwares who can run on GPU and number of agents
> aren't a problem.
>
> Good luck
>
>
>
> 2013/4/30 David García Callejas <david.garcia.callejas at gmail.com>
>
>> H
>> ​i all,
>>
>> I am interested in performing a large-scale (10⁶-10⁸ agents)
>> individual-based simulation in R. I have checked other options (NetLogo,
>> Repast) and they either can't handle such quantities of agents or require
>> deep knowledge of other programming languages. So I would like to know if
>> there are any comprehensive packages optimized to perform IB simulations...
>> I have heard about the "simecol" package, has anyone used it for
>> large-scale simulations?
>>
>> Every suggestion is welcome, thanks
>>
>> David
>>
>>         [[alternative HTML version deleted]]
>>
>>
>> _______________________________________________
>> R-sig-ecology mailing list
>> R-sig-ecology at r-project.org
>> https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
>>
>>


-- 
Julian Mariano Burgos, PhD
Hafrannsóknastofnunin/Marine Research Institute
Skúlagata 4, 121 Reykjavík, Iceland
Sími/Telephone : +354-5752037
Bréfsími/Telefax:  +354-5752001
Netfang/Email: julian at hafro.is