FLAMEGPU is a multi agent simulation framework developed at the University of Sheffield. It uses GPUs to accelerate the simulation of multiagent systems but abstracts the GPU architecture away from users so that they can write models using a high level of abstraction (without having to write GPU code).
To install FLAME GPU you should checkout the latest master branch of the code which has Linux support
git clone https://github.com/FLAMEGPU/FLAMEGPU.git
To run the FLAME GPU examples you will need to be on a GPU node. You can start an interactive session using
qrshx -l gpu=1 -l gpu_arch=nvidia-k40m -l rmem=13G
You will then need to load the relevant modules
module load libs/cuda/8.0.44
module load compilers/gcc/4.9.2
You can now navigate to the FLAME GPU examples folder and build the examples. e.g. to build all the examples in console mode
cd FLAMEGPU/examples
make console
Building the full suite of examples can take a while, instead you may wish to build individual examples. eg. to build a single example in console mode
cd FLAMEGPU/examples/GameOfLife
make console
You can now run the console versions of the example models by navigating to FLAMEGPU/bin/x64 and calling the appropriate shell script. e.g. from the FLAMEGPU/examples/GameOfLife directory
../../bin/linux-x64/GameOfLife_console.sh
or by manually running the executable with the appropriate command line arguments. e.g. from the FLAMEGPU/examples/GameOfLife directory
../../bin/linux-x64/Release_Console/GameOfLife iterations/0.xml 1
FLAME GPU will run the example for one iteration and output the model state to a file 1.xml in the model’s iterations directory.
Visualisation currently does not work with X forwarding as FLAME GPU uses complex rendering techniques which are not supported.
For more information please see the FLAME GPU Documentation. Additional information on compiling FLAME GPU examples can be found via
make help
And additional information on the command line arguments via
../../bin/linux-x64/Release_Console/GameOfLife --help