hybrid computing and ABM

[Marcus G. Daniels]

>
> A very interesting paper about accelerating agent-based models with graphics
> cards (GPU). I placed the paper at
> http://www.friam.org/DSouza_GPUacceleratedABM.pdf

Pretty neat.    The fact is that if we want to take advantage of these
kind of architectures in a general purpose way, we need compiler
technology appropriate for them.   Toolkits may still work if they can
physically live within the constraints of such hardware, but Java VMs
certainly cannot (e.g. a 256KB local store on a Cell SPU).    Better yet
would be to have an ABM language directly integrated with the compiler
and have, e.g.  agent distribution between different memory areas and
processors directly managed.  The compiler is in the best position to
manage hardware.  Certainly better than most modelers.   One problem
with software layering (ABM toolkits) is that compilers can't see
through all that to prove those operations that are safe to do in parallel.

LLVM (http://www.llvm.org/) would be one foundation, and a lot of work
has been done on GCC to make new front ends easier.  

Marcus