The Scalability of JavaSpaces for the Implementation of different Parallel Evolutionary Algorithms
Evolutionary algorithms (EAs) have been successfully deployed in many areas such as medicine, biotechnology, robotics, telecommunications, and economics to mention only a few. However, EAs do not scale up when the problems to be tackled become more complicated. This has given rise to parallel evolutionary algorithms (PEAs). The implementation of parallel applications, however, is not trivial, often requiring specific hardware and software settings and frequently resulting in barely reusable programs. This study investigates the JavaSpaces technology, which is known to simplify the implementation of parallel applications. It is independent of hardware and software settings and allows the implementation of highly reusable applications. We utilised the JavaSpaces technology to implement three PEA paradigms: the synchronous master slave PEA, the asynchronous master slave PEA and the coarse grained PEA. We examine the scalability of the three PEA implementations for a particular problem: the construction of fuzzy classification rules from data.[Full Paper]
For each technical report listed here, copyright and all intellectual property rights remain with the respective authors. Copyright is effective from the year of publication in each case. By downloading a file from this page, you agree to use it only for purposes of research and scholarship. Any other use of this material or storage of it in any medium or its sale or distribution in any form is expressly forbidden without prior written permission from the authors concerned.