A Parallel Genetic Algorithm Framework for Cloud Computing Applications

作者: Elena Apostol , Iulia Băluţă , Alexandru Gorgoi , Valentin Cristea

DOI: 10.1007/978-3-319-13464-2_9

关键词: Parallel genetic algorithmCloud computingEvolutionary algorithmGenetic algorithmGenetic representationTheoretical computer scienceComputer scienceOptimization problem

摘要: Genetic Algorithms (GA) are a subclass of evolutionary algorithms that use the principle evolution in order to search for solutions optimization problems. Evolutionary by their nature very good candidates parallelization, and genetic do not make an exception. Moreover, researchers have stated with larger populations tend obtain better faster convergence. These main reasons why they can benefit from MapReduce implementation. However, research this area is still young, there only few approaches adapting model.

参考文章(8)
Xavier Llorà, Abhishek Verma, Roy H. Campbell, David E. Goldberg, When Huge Is Routine: Scaling Genetic Algorithms and Estimation of Distribution Algorithms via Data-Intensive Computing Studies in computational intelligence. ,vol. 269, pp. 11- 41 ,(2010) , 10.1007/978-3-642-10675-0_2
Parallel and Distributed Computational Intelligence Springer Publishing Company, Incorporated. ,(2010) , 10.1007/978-3-642-10675-0
Carsten Witt, Population size versus runtime of a simple evolutionary algorithm Theoretical Computer Science. ,vol. 403, pp. 104- 120 ,(2008) , 10.1016/J.TCS.2008.05.011
J. E. Beasley, OR-Library: Distributing Test Problems by Electronic Mail Journal of the Operational Research Society. ,vol. 41, pp. 1069- 1072 ,(1990) , 10.1057/JORS.1990.166
Chao Jin, Christian Vecchiola, Rajkumar Buyya, None, MRPGA: An Extension of MapReduce for Parallelizing Genetic Algorithms ieee international conference on escience. pp. 214- 221 ,(2008) , 10.1109/ESCIENCE.2008.78
Di-Wei Huang, Jimmy Lin, Scaling Populations of a Genetic Algorithm for Job Shop Scheduling Problems Using MapReduce ieee international conference on cloud computing technology and science. pp. 780- 785 ,(2010) , 10.1109/CLOUDCOM.2010.18