Accelerating Cellular Automata Evolution on Graphics Processing Units

Abstract—As design of cellular automata rules using conventional methods is a difficult task, evolutionary algorithms are often utilized in this area. However, in that case, high computational demands need to be met. This problem may be partially solved by parallelization. Since parallel supercomputers and server clusters are expensive and often overburdened, this paper proposes the evolution of cellular automata rules on small and inexpensive graphic processing units. The main objective of this paper is to demonstrate that evolution of cellular automata rules can be accelerated significantly using graphics processing units. Several methods of speeding-up the evolution of cellular automata rules are proposed, evaluated and compared, some with very good results. Also a comparison is made between mid-end and high-end graphics accelerator card based on the results of evolution speedup. The proposed methods are evaluated using two benchmark problems. Keywords–cellular automata; parallel computing; GPU; CUDA; genetic alghorith