Title :
A CUDA-Based Multi-Channel Particle Swarm Algorithm
Author :
Li Wenna ; Zhang Zhenyu
Author_Institution :
Sch. of Electr. Eng. & Autom., Tianjin Polytech. Univ., Tianjin, China
Abstract :
In order to improve the efficiency and accuracy of the particle swarm optimization algorithm in optimization, the parallel CUDA-based particle swarm algorithm is proposed and developed. With the Compute Unified Device Architecture (CUDA) technology, the parallel data structure is defined, and the mechanism of computing tasks mapping to CUDA is described. From the optimization experiments results of 4 benchmark functions it shows that the CUDA-based parallel algorithm can greatly save computing time and improve computing accuracy. This new PSO algorithm is more suitable for the relevant application of the particle swarm algorithm.
Keywords :
data structures; parallel processing; particle swarm optimisation; CUDA based multichannel particle swarm algorithm; CUDA based parallel algorithm; Compute Unified Device Architecture; PSO algorithm; efficiency improvement; parallel CUDA based particle swarm algorithm; parallel data structure; particle swarm optimization algorithm; Algorithm design and analysis; Computer architecture; Graphics processing unit; Instruction sets; Kernel; Optimization; Particle swarm optimization;
Conference_Titel :
Control, Automation and Systems Engineering (CASE), 2011 International Conference on
Conference_Location :
Singapore
Print_ISBN :
978-1-4577-0859-6
DOI :
10.1109/ICCASE.2011.5997786
