Affiliation:
1. Department of Computer Science and Computer Engineering, La Trobe University, Bundoora, Australia
Abstract
Large-scale Agent-Based Modelling and Simulation (ABMS) is a field of research that is becoming increasingly popular as researchers work to construct simulations at a higher level of complexity and realism than previously done. These systems can not only be difficult and time consuming to implement, but can also be constrained in their scope due to issues arising from a shortage of available processing power. This work simultaneously presents solutions to these two problems by demonstrating a model for ABMS that allows a developer to design their own simulation, which is then automatically converted into code capable of running on a mainstream Graphical Processing Unit (GPU). By harnessing the extra processing power afforded by the GPU this paper creates simulations that are capable of running in real-time with more autonomous agents than allowed by systems using traditional x86 processors.
Reference32 articles.
1. Chan, W. K. V., Son, Y.-J., & Macal, C. M. (2010). Agent-based simulation tutorial - simulation of emergent behavior and differences between agent-based simulation and discrete-event simulation. In Proceedings of the Winter Simulation Conference, ser. WSC ’10. Winter Simulation Conference (pp. 135–150). Available at http://dl.acm.org/citation.cfm?id=2433508.2433525
2. To GPU synchronize or not GPU synchronize?
3. Goodnight, N., Woolley, C., Lewin, G., Luebke, D., & Humphreys, G. (2003). A multigrid solver for boundary value problems using programmable graphics hardware. In Proceedings of the ACM SIGGRAPH/EUROGRAPHICS Conference on Graphics Hardware (ser. HWWS ’03) (pp. 102–111). Aire-la-Ville, Switzerland, Switzerland: Eurographics Association; Available at http://dl.acm.org/citation.cfm?id=844174.844190
4. Efficient implementation of GPGPU synchronization primitives on CPUs
5. Parallel Simulation of Large-Scale Artificial Society on CPU/GPU Mixed Architecture