摘要
提出了一种基于GPU加速的细粒度并行粒子群算法,将并行PSO求解过程转化为GPU纹理渲染过程,使PSO算法在GPU中加速执行,在取得了较好的优化效果的同时,增大了细粒度并行的粒子规模,提高了算法的运算速度,并为普通用户的并行PSO工作提供了一种可行的方法.
We raised a fine-grained PSO algorism based on GPU acceleration, which converts the process of working-out into the process of texture-rendering based on GPU, making PSO greatly accelerated in it. As a-chieving a good optimization effect, it also increases the particle population in the fine-grained parallelism, speeds up its running and provides ordinary user with a feasible PSO solution.
出处
《哈尔滨工业大学学报》
EI
CAS
CSCD
北大核心
2006年第12期2162-2166,共5页
Journal of Harbin Institute of Technology
基金
国家自然科学基金资助重点项目(701710407003102070571009)
高等学校博士点基金资助项目(20010141025)
教育部重点科研项目(03052)
关键词
粒子群优化算法
并行处理
GPU
细粒度
particle swarm optimization algorithm
parallel process
GPU
fine-grained
