
© Springer International Publishing Switzerland 2015
H. Handa et al. (eds.), Proc. of the 18th Asia Pacific Symp. on Intell. & Evol. Systems – Vol. 1,
55
Proceedings in Adaptation, Learning and Optimization 1, DOI: 10.1007/978-3-319-13359-1_5
Parallel Particle Swarm Optimization Using Message
Passing Interface
*
Guang-Wei Zhang
1,2,3
, Zhi-Hui Zhan
1,2,3,**
, Ke-Jing Du
4
, Ying Lin
2,3,5
,
Wei-Neng Chen
2,3,4
, Jing-Jing Li
6
, and Jun Zhang
1,2,3,4
1
Department of Computer Science, Sun Yat-sen University, China, 510275
2
Key Lab. Machine Intelligence and Advanced Computing, Ministry of Education, China
3
Engineering Research Center of Supercomputing Engineering Software, MOE, China
4
School of Advanced Computing, Sun Yat-sen University, China, 510275
5
Department of Psychology, Sun Yat-sen University, China, 510275
6
School of Computer Science, South China Normal University, China
zhanzhh@mail.sysu.edu.cn
Abstract. Parallel computation is an efficient way to combine the advantages of
different computation paradigms to obtain promising solution. In order to analyze
the performance of parallel computation techniques to the particle swarm optimiza-
tion (PSO) algorithm, a parallel particle swarm optimization (PPSO) is proposed in
this paper.
Since the theorem of “no free lunch” exists, there is not an optimization
algorithm that can perfectly tackle all problems. The PPSO provides a paradigm to
combine different variants of PSO algorithms by using the Message Passing Inter-
face (MPI) so that the advantages of diverse PSO algorithms can be utilized. The
PPSO divides the whole evolution process into several stages. At the interval be-
tween two successive stages, each PSO algorithm exchanges the achievement of
their evolution and then continues with the next stage of evolution. By merging the
global model PSO (GPSO), the local model PSO (LPSO), the bare bone PSO
(BPSO), and the comprehensive learning PSO (CLPSO), the PPSO achieves higher
solution quality than the serial version of these four PSO algorithms, according to
the simulation results on benchmark functions.
Keywords: Parallel particle swarm optimization (PPSO), evolutionary algo-
rithm, evolution stage, Message Passing Interface (MPI).
1 Introduction
Particle swarm optimization (PSO) [1], introduced by Kennedy and Eberhart, was
inspired by the bird flocking and fish schooling. The PSO system is made up of a
group of individuals (particles). These particles fly through the search domain while
pursuing optimal solution to the problem. During the flying, each particle remembers
*
This work was supported in part by the National High-Technology Research and Develop-
ment Program (863 Program) of China No.2013AA01A212, in part by the National Natural
Science Fundation of China (NSFC) with No. 61402545, 61332002, and 61300044, and in
part by the NSFC for Distinguished Young Scholars with No. 61125205.
**
Corresponding author.