This paper describes the design of an agent-based modelling framework for high performance computing. Rather than a collection of methods that require parallel programming expertise the framework presented allows modellers to concentrate on the model while the framework handles the efficient execution of simulations. The framework uses a state machine based representation of agents that allows a statically calculated optimal ordering of agent execution and parallel communication routines. Some experiments with the current implementation and the results of using a simple communication dominant model for benchmarking performance are reported. The model with half a million agents is used to show that a parallel efficiency of above 80% is achievable when distributed over 432 processors. Future improvements are discussed including data dependency analysis, vector operations over agents, and dynamic task scheduling.
2012 IEEE 14th International Conference on High Performance Computing and Communication 2012 IEEE 9th International Conference on Embedded Software and Systems
%0 Conference Paper
%1 coakley2012exploitation
%A Coakley, S.
%A Gheorghe, M.
%A Holcombe, M.
%A Chin, S.
%A Worth, D.
%A Greenough, C.
%B 2012 IEEE 14th International Conference on High Performance Computing and Communication 2012 IEEE 9th International Conference on Embedded Software and Systems
%D 2012
%K agent-based_simulation simulation software
%P 538-545
%R 10.1109/HPCC.2012.79
%T Exploitation of High Performance Computing in the FLAME Agent-Based Simulation Framework
%U http://ieeexplore.ieee.org/abstract/document/6332218/?reload=true
%X This paper describes the design of an agent-based modelling framework for high performance computing. Rather than a collection of methods that require parallel programming expertise the framework presented allows modellers to concentrate on the model while the framework handles the efficient execution of simulations. The framework uses a state machine based representation of agents that allows a statically calculated optimal ordering of agent execution and parallel communication routines. Some experiments with the current implementation and the results of using a simple communication dominant model for benchmarking performance are reported. The model with half a million agents is used to show that a parallel efficiency of above 80% is achievable when distributed over 432 processors. Future improvements are discussed including data dependency analysis, vector operations over agents, and dynamic task scheduling.
@inproceedings{coakley2012exploitation,
abstract = {This paper describes the design of an agent-based modelling framework for high performance computing. Rather than a collection of methods that require parallel programming expertise the framework presented allows modellers to concentrate on the model while the framework handles the efficient execution of simulations. The framework uses a state machine based representation of agents that allows a statically calculated optimal ordering of agent execution and parallel communication routines. Some experiments with the current implementation and the results of using a simple communication dominant model for benchmarking performance are reported. The model with half a million agents is used to show that a parallel efficiency of above 80% is achievable when distributed over 432 processors. Future improvements are discussed including data dependency analysis, vector operations over agents, and dynamic task scheduling.},
added-at = {2017-09-07T08:00:37.000+0200},
author = {Coakley, S. and Gheorghe, M. and Holcombe, M. and Chin, S. and Worth, D. and Greenough, C.},
biburl = {https://www.bibsonomy.org/bibtex/268b27f71c813ac2c0474a7b5e6922b9c/peter.ralph},
booktitle = {2012 IEEE 14th International Conference on High Performance Computing and Communication 2012 IEEE 9th International Conference on Embedded Software and Systems},
doi = {10.1109/HPCC.2012.79},
interhash = {4a6db22e07cbc63a72310eeaed1ac7cb},
intrahash = {68b27f71c813ac2c0474a7b5e6922b9c},
keywords = {agent-based_simulation simulation software},
month = {June},
pages = {538-545},
timestamp = {2017-09-07T08:00:37.000+0200},
title = {Exploitation of High Performance Computing in the {FLAME} Agent-Based Simulation Framework},
url = {http://ieeexplore.ieee.org/abstract/document/6332218/?reload=true},
year = 2012
}