Massively Parallel Genetic Algorithm - Pattern Search for Nonlinear Optimization with Graphics Hardware Acceleration

Abstract

This paper presents a massively parallel Genetic Algorithm – Pattern Search (GA-PS) with graphics hardware acceleration on bound constrained nonlinear optimization problems. The objective of this study is to determine the effectiveness of using Graphics Processing Units (GPU) as a hardware platform for Genetic Algorithms (GA). The global search of the GA is enhanced by a local Pattern Search (PS) improvement phase. The hybrid GA-PS method is implemented in the GPU environment and compared to a similar implementation in the common computing environment with a Central Processing Unit (CPU). Computational results indicate that GPU-accelerated GA-PS method is orders of magnitude faster than the corresponding CPU implementation. The main contribution of this paper is the parallelization analysis and performance analysis of the hybrid GA-PS with GPU acceleration. The computational results demonstrate the potential of using GPU hardware for parallel massively optimization on a personal computer.