Differential Evolution: A Practical Approach to Global Optimization (Natural Computing Series)
Differential Evolution: A Practical Approach to Global Optimization (Natural Computing Series)
MapReduce: simplified data processing on large clusters
OSDI'04 Proceedings of the 6th conference on Symposium on Opearting Systems Design & Implementation - Volume 6
MRPGA: An Extension of MapReduce for Parallelizing Genetic Algorithms
ESCIENCE '08 Proceedings of the 2008 Fourth IEEE International Conference on eScience
Distributed simulated annealing with mapreduce
EvoApplications'12 Proceedings of the 2012t European conference on Applications of Evolutionary Computation
Concurrent differential evolution based on generational model for multi-core CPUs
SEAL'12 Proceedings of the 9th international conference on Simulated Evolution and Learning
High performance parallel evolutionary algorithm model based on MapReduce framework
International Journal of Computer Applications in Technology
Hi-index | 0.00 |
MapReduce is a promising programming model for developing distributed applications due to its superb simplicity, scalability and fault tolerance. This paper demonstrates how to apply MapReduce and the open source Hadoop framework for a quick and easy parallelization of the Differential Evolution algorithm. Instead of parallelizing the whole evolution process, our simple solution is to only apply the MR model to the fitness evaluation part, which usually consumes most of the running time. Two alternative approaches are investigated, i.e., population based and data based. Experimental results reveal that even though the population based approach is a better way, the extra cost of Hadoop DFS I/O operations and system bookkeeping overhead significantly reduces the benefits of parallelism.