Specialized Multicore Architectures Supporting Efficient Multi-Agent Simulations

Two new multiprocessor architectures to accelerate the simulation of multi-agent systems based on the massively parallel GCA (Global Cellular Automata) model are presented. The GCA model is suited to describe and simulate different multi-agent systems. The designed and implemented architectures mainly consist of a set of processors (NIOS II) and a network. The multiprocessor systems allow the implementation in a flexible way through programming, thus simulating different behaviors on the same architecture. Two architectures, one with up to 16 processors, were implemented on an FPGA. The first architecture uses hardware hash functions in order to reduce the overall simulation time, but lacks scalability. The second architecture uses an agent memory and a cell field memory. This improves the scalability and further increases the performance.

[1]  Christian Schäck,et al.  Network Optimization of a Multiprocessor Architecture for the MassivelyParallel Model GCA , 2009 .

[2]  Rolf Hoffmann,et al.  A scalable configurable architecture for the massively parallel GCA model , 2008, IPDPS.

[3]  Stefan Waldschmidt,et al.  GCA: Global Cellular Automata. A Flexible Parallel Model , 2001, PaCT.

[4]  Michael Schreckenberg,et al.  A cellular automaton model for freeway traffic , 1992 .

[5]  Kai Nagel,et al.  Multi-agent traffic simulation with CUDA , 2009, 2009 International Conference on High Performance Computing & Simulation.

[6]  R. D'Souza.  SUGARSCAPE ON STEROIDS : SIMULATING OVER A MILLION AGENTS , 2007 .

[7]  Christian Schäck,et al.  GCA Multi-Softcore Architecture for Agent Systems Simulation , 2009, GI Jahrestagung.

[8]  Ronald Thenius,et al.  Kollektive Sammel-Entscheidungen: Eine Multi-Agenten-Simulation einer Honigbienenkolonie , 2006 .

[9]  Stefan Waldschmidt,et al.  Global Cellular Automata GCA: An Universal Extension of the CA Model , 2000 .

[10]  Rolf Hoffmann,et al.  Generated Horizontal and Vertical Data Parallel GCA Machines for the N-Body Force Calculation , 2009, ARCS.

[11]  Jiming Liu,et al.  A Massively Multi-agent System for Discovering HIV-Immune Interaction Dynamics , 2004, MMAS.

[12]  Christian Schäck,et al.  A Multiprocessor Architecture with an Omega Network for the Massively Parallel Model GCA , 2009, SAMOS.

[13]  Christian Schäck,et al.  Efficient traffic simulation using the GCA model , 2010, 2010 IEEE International Symposium on Parallel & Distributed Processing, Workshops and Phd Forum (IPDPSW).

[14]  Kai Nagel,et al.  Using common graphics hardware for multi-agent traffic simulation with CUDA , 2009, SIMUTools 2009.