Parallel computing of GAME models. Pavel Kordik, Jakub Spirk, Ivan Simecek

Abstract. With recent development of multi-core and multi-processor computers, single thread algorithms use just fraction of possible computing resources available on single personal computer. The trend is to develop distributed versions of algorithms so they can run on several cores in parallel efficiently using all resources available. In this paper we present an efficient distributed version of the GAME algorithm for inductive models evolution. We also discuss the possibilities and assets of parallelizing evolution of inductive models. Our experimental results demonstrate that for two core processors the distributed GAME algorithm achieves 1.7 speedup in average against the serial version. For eight cores, the speedup is 3.5 in average.

Keywords. Inductive modelling, distributed computing, parallel processing, FAKE GAME.

References.
1.  http://en.wikipedia.org/wiki/symmetricmultiprocessing, August 2008.
2. P. Kordik. Fully Automated Knowledge Extraction using Group of Adaptive Models Evolution. PhD thesis, Czech Technical University in Prague, FEE, Dep. of Comp. Sci. and Computers, FEE, CTU Prague, Czech Republic, September 2006.
3. O. Koshulko and A. Koshulko. Adaptive parallel implementation of the combinatorial gmdh algorithm. In IWIM, Prague, 2007.
4. S. Krishnaprasad. Uses and abuses of amdahl's law. J. Comput. Small Coll., 17(2):288-293, 2001.
5. S. W. Mahfoud. Niching methods for genetic algorithms. Technical Report 95001, Illinois Genetic Algorithms Laboratory (IlliGaL), University of Ilinios at Urbana-Champaign, May 1995.
6. S. Oaks and H. Wong. Java Threads, Second Edition. O'Reilly & Associates, Inc., Sebastopol, CA, USA, 1999.
7. G. G. Robertson. Parallel implementation of genetic algorithms in a classifier system. In Proceedings of the Second International Conference on Genetic Algorithms on Genetic algorithms and their application, pages 140-147, Mahwah, NJ, USA, 1987. Lawrence Erlbaum Associates, Inc.