Abstract
Generally evolutionary algorithms designed for parallel environments improve their execution time. However, if the algorithm has different individual evaluation costs, or if it is executed under multiprogramming or in heterogeneous platforms, performance can be adversely affected. This paper presents the integration of a load balancing scheme designed in order to improve the execution time of a parallel evolutionary algorithm. This scheme dynamically balances the load by transferring data from overloaded nodes to underloaded nodes. Data migration is relatively simple, both because the application data is represented as a list of individuals, and because each individual is evaluated independently. The balance mechanism is integrated by means of a single call, contrary to other works where the application code and the balance service are mixed. This proposal can be extended and integrated into any application with similar data representation and management. It is shown how under certain system conditions, the approach offers good performance.
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 subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Adamidis, P.: Parallel evolutionary algorithms: A review, Web Page, http://citeseer.nj.nec.com/17679.html
Barak, A.: Official mosix web, http://www.mosix.cs.huji.ac.il/
Allen, M., Wilkinson, B.: Parallel Programming, Techniques and Applications Using Networked Worstations and Parallel Computers. Prentice-Hall, Englewood Cliffs (1999)
Bubak, M., Sowa, K.: Parallel object-oriented library of genetic algorithms. pp. 12 (1996)
Buyya, R.: High Performance Cluster Computing: Architectures and Systems, vol. 1. Prentice-Hall, Englewood Cliffs (1999)
Buyya, R.: High Performance Cluster Computing: Programming and Applications, vol. 2. Prentice-Hall, Englewood Cliffs (1999)
Román Alonso, G.: Contribution a l’etude des algorithmes d’allocation dynamique sur des machines MIMD a memoire distribuee. PhD thesis, Universite de Technologie de Compiegne France (1997)
Williams, J., Riessen, G.A., Yao, X.: Pepnet: Parallel evolutionary programming for constructing artificial neural networks (1997)
Huang, M.-C., Hossein, S.: Load balancing in computer networks. In: International Society for Computers and Their Applications ISCA (2002)
Merelo Guervós, J.: Algoritmos Evolutivos en Perl (September 2002)
Castro García, M.A.: Balance de carga en un sistema paralelo con memoria distribuida. Master’s thesis, CINVESTAV IPN. Computer Section (May 2001)
Goddard Close, J., Martínez Licona, A.E.: Definición de una red neuronal para clasificación por medio de un programa evolutivo. Revista Mexicana de Ingenier ía Biomédica XXII(1), 4–11 (2001)
Goddard Close, J., Fernandez Trejo, O., Martínez Licona, A.E., Román Alonso, G.: Alternativa para la solución del problema de clasificación por medio de un programa evolutivo. In: XXV Congreso Nacional de Ingeniería Biomédica, Monterrey N. L. México (2001)
Patterson, D.W.: Artificial Neural Networks: Theory and Appliations, 1st edn. Prentice-Hall, Englewood Cliffs (1998)
Hart, P.E., Duda, R.O.: Pattern Classification, 2nd edn. Wiley, John and Sons, Chichester (1999) (incorporated)
MPI The Message Passing Interface Standard, Web Page, http://www.netlib.org/mpi
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Castro, M., Román, G., Buenabad, J., Martínez, A., Goddard, J. (2004). Integration of Load Balancing into a Parallel Evolutionary Algorithm. In: Ramos, F.F., Unger, H., Larios, V. (eds) Advanced Distributed Systems. ISSADS 2004. Lecture Notes in Computer Science, vol 3061. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-25958-9_20
Download citation
DOI: https://doi.org/10.1007/978-3-540-25958-9_20
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-22172-2
Online ISBN: 978-3-540-25958-9
eBook Packages: Springer Book Archive