GPU-based Acceleration of the Genetic Algorithm

Genetic algorithm (GA) is a stochastic optimization method inspired by nature evolution. Because of their parallel nature, they have been parallelized many times. Graphic Processing Units (GPU) were originally targeted for rasterization of graphics primitives. Today GPU’s are more likely fast multicore processors capable of performing complex mathematical tasks. There are many ways how to exploit GPU’s potential for general purpose computation (GPGPU). One option is to employ Compute Unified Device Architecture (CUDA) framework.