计算机科学 ›› 2014, Vol. 41 ›› Issue (9): 263-268.doi: 10.11896/j.issn.1002-137X.2014.09.050
陈风,田雨波,杨敏
CHEN Feng,TIAN Yu-bo and YANG Min
摘要： 应用图形处理器(GPU)来加速粒子群优化(PSO)算法并行计算时,为突出其加速性能,经常有文献以恶化CPU端PSO算法性能为代价。为了科学比较GPU-PSO算法和CPU-PSO算法的性能,提出用“有效加速比”作为算法的性能指标。文中给出的评价方法不需要CPU和GPU端粒子数相同,将GPU并行算法与最优CPU串行算法的性能作比较,以加速收敛到目标精度为准则,在统一计算设备架构(CUDA)下对多个基准测试函数进行了数值仿真实验。结果表明,在GPU上大幅增加粒子数能够加速PSO算法收敛到目标精度,与CPU-PSO相比,获得了10倍以上的“有效加速比”。
| [1] Kennedy J,Eberhart R．Particle swarm optimization [C]∥Proceedings of the IEEE International Conference on Neural Networks．Perth,WA,1995,4:1942-1948 [2] Poli R,Kennedy J,Blackwell T．Particle swarm optimization:an overview [J]．Swarm Intelligence,2007,1(1):33-57 [3] Singhal G,Jain A,Patnaik A．Parallelization of particle swarmoptimization using message passing interfaces (MPIs) [C]∥IEEE World Congress on Nature & Biologically Inspired Computing．Coimbatore,2009:67-71 [4] Deep K,Sharma S,Pant M．Modified parallel particle swarm optimization for global optimization using Message Passing Interface [C]∥2010 IEEE Fifth International Conference on Bio-Inspired Computing:Theories and Applications．Changsha,2010:1451-1458 [5] Wang D Z,Wu C H,et al．Parallel multi-population ParticleSwarm Optimization Algorithm for the Uncapacitated Facility Location problem using OpenMP [C]∥IEEE Congress on Evolutionary Computation．HK,2008:1214-1218 [6] Venayagamoorthy G K,Gudise V G．Swarm intelligence for digital circuits implementation on field programmable gate arrays platforms [C]∥Proceedings of the IEEE Conference on Evolvable Hardware．2004:83-86 [7] Maeda Y,Matsushita N．Simultaneous Perturbation ParticleSwarm Optimization Using FPGA [C]∥IEEE International Joint Conference on Neural Networks．Orlando,FL,2007:2695-2700 [8] Veronese L,Krohling R．Swarm’s flight:Accelerating the particles using C-CUDA [C]∥Proceedings of the IEEE Congress on Evolutionary Computation．Trondheim,2009:3264-3270 [9] Calazan R M,Nedjah N,de Macedo Mourelle L．Parallel GPU-based implementation of high dimension Particle Swarm Optimizations [C]∥2013 IEEE Fourth Latin American Symposium on Circuits and Systems．Cusco,2013:1-4 [10] Zhou Y,Tan Y．GPU-based parallel particle swarm optimization [C]∥Proceedings of the IEEE Congress on Evolutionary Computation．Trondheim,2009:1493-1500 [11] Venneschi L,Codecasa D,Mauri G．An empirical comparison of parallel and distributed particle swarm optimization methods [C]∥Proceedings of the 12th annual conference on Genetic and evolutionary computation．Portland,Oregon,2010:15-22 [12] Solomon S,Thulasiraman P,Thulasiram R．Collaborative multi-swarm PSO for task matching using graphics processing units [C]∥Proceedings of the 13th annual conference on Genetic and evolutionary computation．Dublin,Ireland,2011:12-16 [13] Mussi L,Daolio F,Cagnoni S．Evaluation of parallel particleswarm optimization algorithms within the CUDA architecture [J]．Information Sciences,2010,181(20):4642-4657 [14] Mussi L,Nashed Y S G,Cagnoni S．GPU-based asynchronousparticle swarm optimization [C]∥Proceedings of the 13th annual conference on Genetic and evolutionary computation．Dublin,Ireland,2011:1555-1562 [15] 张庆科,杨波,王琳,等．基于GPU的现代并行优化算法 [J]．计算机科学,2012,39(4):304-311 [16] 蔡勇,李光耀,王琥．基于CUDA的并行粒子群优化算法的设计与实现 [J]．计算机应用研究,2013,30(8):2415-2418 [17] Roberge V,Tarbouchi M．Efficient parallel Particle Swarm Optimizers on GPU for real-time harmonic minimization in multilevel inverters [C]∥38th Annual Conference on IEEE Industrial Electronics Society．Montreal,2012:2275-2282 [18] Bastos-Filho C,Oliveira M,Nascimento D,et al．Impact of theRandom Number generator quality on particle swarm optimization algorithm running on graphic processor units [C]∥IEEE 10th International Conference on Hybrid Intelligent Systems．Atlanta,2010:85-90 [19] Shi Y,Eberhart R．A Modified Particle Swarm Optimizer [C]∥Proceedings of the IEEE International Conference on Evolutionary Computation．Anchorage,AK,1998:69-73 |
| No related articles found! |
|
||
首页