| 三种GPU并行的自适应邻域模拟退火算法 |
| |
| 引用本文： | 林 敏,钟一文. 三种GPU并行的自适应邻域模拟退火算法[J]. 计算机工程与应用, 2015, 51(22): 70-76 |
| |
| 作者姓名： | 林 敏 钟一文 |
| |
| 作者单位： | 福建农林大学 计算机与信息学院，福州 350002 |
| |
| 摘 要： | 提出了三种新的GPU并行的自适应邻域模拟退火算法，分别是GPU并行的遗传-模拟退火算法，多条马尔可夫链并行的退火算法，基于BLOCK分块的GPU并行模拟退火算法，并通过对GPU端的程序采取合并内存访问，避免bank冲突，归约法等方式进一步提升了性能。实验中选取了11个典型的基准函数，实验结果证明这三种GPU并行退火算法比nonu-SA算法具有更好的精度和更快的收敛速度。
|
| 关 键 词： | 图形处理器（GPU） 遗传算法 自适应邻域 计算统一设备架构（CUDA） Guassion分布 |
Three GPU-based parallel simulated annealing algorithm with adaptive neighborhood |
| |
| LIN Min,ZHONG Yiwen. Three GPU-based parallel simulated annealing algorithm with adaptive neighborhood[J]. Computer Engineering and Applications, 2015, 51(22): 70-76 |
| |
| Authors: | LIN Min ZHONG Yiwen |
| |
| Affiliation: | College of Computer and Information, Fujian Agriculture and Forestry University, Fuzhou 350002, China |
| |
| Abstract: | Three new GPU-based parallel simulated annealing algorithms with adaptive neighborhood are proposed in this paper. They are parallel genetic-simulated annealing algorithm based on GPU, parallel annealing algorithm with multiple Markov chains, and parallel annealing algorithm based on block. Several novel strategies adopted in these algorithms such as coalescent memory access, avoiding bank conflict, and reduction improve the performance. The experiments tested on 11 typical benchmark functions show the new three algorithms have better accuracy and faster convergence speed than the nonu-SA algorithm. |
| |
| Keywords: | Graphic Processing Unit（GPU） genetic algorithm adaptive neighborhood Compute Unified Device Architecture（CUDA） Gaussian distribution |
|
| 点击此处可从《计算机工程与应用》浏览原始摘要信息 |
|
点击此处可从《计算机工程与应用》下载免费的PDF全文 |
