|
From: North, M. <no...@an...> - 2008-02-01 14:58:16
|
Dan: =20 100 million to 1 billion is a lot of agents! It depends on the = complexity of the agents themselves, but you usually need a rather large = computer to meaningfully model this type of thing in any form. I = certainly would not recommend starting at this scale! You might start = with Repast using smaller numbers of agents then move towards = increasingly customized code as you scale up. =20 In terms of agent locations, it again depends on what you intend. = Broadly speaking, models only require explicit spatial locations for = components if the agent behavior or output analysis depends on explicit = spatial locations. For example, modelers sometimes use networks in place = of explicit spatial locations since for some models only the relative = connectivity is important, as opposed to exact lattitude and longitude. =20 Mike ________________________________ From: rep...@li... on behalf of = ltdanp21 Sent: Thu 1/31/2008 3:49 PM To: rep...@li... Subject: [Repast-developer] Can Repast be used to build large scale ABMs = (1 billion agents)? 2 questions about building large-scale ABMs with Repast: Can Repast be used to build models with 100 million to 1 billion agents (think India)? Can Repast build ABMs where the location of every single school, house, = and work are not explicitly known (clearly not possible in the case of a = region as big as India). Dan -- View this message in context: = http://www.nabble.com/Can-Repast-be-used-to-build-large-scale-ABMs-%281-b= illion-agents%29--tp15215597p15215597.html Sent from the repast-developer mailing list archive at Nabble.com. -------------------------------------------------------------------------= This SF.net email is sponsored by: Microsoft Defy all challenges. Microsoft(R) Visual Studio 2008. http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/ _______________________________________________ Repast-developer mailing list Rep...@li... https://lists.sourceforge.net/lists/listinfo/repast-developer |