Abstract
In order to improve the efficiency of hybrid particle swarm optimization (PSO) algorithm, a PSO merging simulated annealing and hill climbing (SAHCPSO) is implemented based on a three-level parallel model to increase its convergence speed and to decrease the operation time. SAHCPSO can enhance the diversity of the population and avoid population premature convergence. By analyzing and optimizing the SAHCPSO, we complete the task mapping on the model and make full use of CPU/GPU heterogeneous cluster resources. Optimization for parallel accessing further improves the efficiency of the algorithm. The parallel SAHCPSO implements the coarse-grained parallelism between computation nodes and fine-grained parallelism within each node, greatly reducing the operation time. The experimental results show that with the increase of particle scale, higher speedup can be obtained. The high efficiency of the parallel strategy of the model makes the parallel SAHCPSO more easily to solve large-scale problems.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Kennedy, J., & Eberhart R. (1995). Particle swarm optimization (pp. 1942–1948). IEEE International Conference on Neural Networks, IEEE, Perth.
Liang, X. M., Dong, S. H., Long, W., & Xiao X. F. (2011). PSO algorithm with dynamical inertial weight vector and dimension mutation. Computer Engineering and Applications, 47(5), 29–31 (In Chinese).
Tian, G., & Lu. F. S. (2011). Measuring MPI/OpenMP+CUDA high-performance computing environment configuration and application. Silicon Valley, 17(9), 118–119 (In Chinese).
Li, J. M., Wan, D. L., Chi, Z. X., & Hu X. P. (2006). A parallel particle swarm optimization algorithm based on fine-grained model with GPU-accelerating. Journal of Harbin Institute of Technology, 12(38), 2162–2836 (In Chinese).
Xu, Q. Y., & Chen Q. K. (2010). Research and implementation of MPI+CUDA model based on SMP clusters. Computer Engineering and Design, 15(31), 3408–3412 (In Chinese).
Liu, Q. K., Ma, M. W., & Yan W. C. (2011). Parallel matrix multiplication based on MPI + CUDA asynchronous model. Journal of Computer Applications, 12(31), 3327–3330 (In Chinese).
Teng, R. D., & Liu Q. K. (2010). Mixed CUDA, MPI and OpenMP in three mode parallel programming. Microcomputer Applications, 9(31), 63–69 (In Chinese).
Sanders, J., & Kandrot E. (2011). CUDA by example: An introduction to general-purpose GPU programming (pp. 162–165). NJ: Addison-Wesley.
Pacheco P. S. (2011). An introduction to parallel programming (pp. 60–62). Burlington: Morgan Kaufmann.
Kirk, D. B., & Hwu W.W. (2012). Programming massively parallel processors: A hands-on approach (2nd ed., pp. 135–137). Burlington: Morgan Kaufmann.
Acknowledgements
This work has been supported by The National Natural Science Foundation of China under research project 41264005 and also supported by The Guangxi department of education under the research project 201102ZD018. We also thank master degree candidate Jiaxing You for his generous help to our work especially in SAHCPSO programming.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Xiao, Y., Liu, Y. (2014). The Implementation of a Hybrid Particle Swarm Optimization Algorithm Based on Three-Level Parallel Model. In: Wong, W.E., Zhu, T. (eds) Computer Engineering and Networking. Lecture Notes in Electrical Engineering, vol 277. Springer, Cham. https://doi.org/10.1007/978-3-319-01766-2_22
Download citation
DOI: https://doi.org/10.1007/978-3-319-01766-2_22
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-01765-5
Online ISBN: 978-3-319-01766-2
eBook Packages: EngineeringEngineering (R0)