Parallelizing Evolutionary Algorithms for Clustering Data
In the paper the problem of using an evolutionary algorithm to partition a dataset into a known number of clusters is considered. A novel approach, based on data decomposition, for parallel computing of the fitness function is proposed. Both the learning set and the population of the evolutionary algorithm are distributed among processors. Processors form a pipeline using the ring topology. In a single step each processor computes the local fitness of its current subpopulation while sending the previous subpopulation to the successor and receiving next subpopulation from the predecessor. Thus it is possible to overlap communication and computation using non-blocking MPI routines. Our approach to parallel fitness computation was applied to differential evolution algorithm. The results of initial experiments show, that for large datasets the algorithm is capable of achieving very good scalability.
KeywordsEvolutionary Algorithm Cluster Problem Trial Population Wide Area Network Trial Vector
Unable to display preview. Download preview PDF.
- 11.Paterlini, S., Krink, T.: High performance clustering with differential evolution. In: Proceedings of the Sixth Congress on Evolutionary Computation (CEC-2004), pp. 2004–2011. IEEE Press, Los Alamitos (2004)Google Scholar
- 12.Price, K., Storn, R.: Minimizing the real functions of the ICEC 1996 optimization contest by differential evolution. In: IEEE International Conference on Evolutionary Computation ICEC 1996, pp. 842–844 (1996)Google Scholar
- 13.Snir, M., Otto, S., Huss-Lederman, S., Walker, D., Dongarra, J.: MPI: The Complete Reference. MIT Press, Cambridge (1996)Google Scholar
- 15.Wyrzykowski, R., Meyer, N., Stroiński, M.: CLUSTERIX: National cluster of Linux systems. In: Across GRIDS 2004, Nicosia, Cyprus (2004)Google Scholar