Abstract

Agent-based modelling and simulation is still an upcoming approach for microsimulation. But a large num-ber of agents with advanced dynamics and interactions requires sophisticated algorithms and lots of computational effort. We try to implement a rather simple but special agent-based model model on GPU-architectures (graphics processing unit). This contribution presents the GPU implementations and and investigates its practicability. Results show that implementation of agent-based systems on a GPU-architecture can deliver enormous speed-ups. Depending on the detailed structure of the system highly advanced algorithms deliver another great performance-boost. The combination of both approaches can deliver up to 1.000 times faster execution of the same simulation.