Exploring GPGPU Acceleration of Process-Oriented Simulations

Abstract. This paper reports on our experiences of using commodity GPUs to speed-up the execution of fine-grained concurrent simulations. Starting with an existing process-oriented ‘boids ’ simulation, we explore a variety of techniques aimed at im-proving performance, gradually refactoring the original code. Successive improve-ments lead to a 10-fold improvement in performance, which we believe can still be im-proved upon, allowing us to explore simulations with larger numbers of agents (30,000 rather than 2,000) interactively and without significant performance degradation