Abstract
The Genetic Algorithm has been used for optimization problems in many areas. One of the attractive features of the Genetic Algorithm is that it lends itself very well to parallel and distributed processing. This feature of the Genetic Algorithm is used in this paper for improving its performance for large and complex optimization problems by implementing it in a distributed environment. The key attribute of the distributed implementation is that it can be used for different types of optimization problems without any modifications. In addition, the Distributed Genetic Algorithm implementation provides fault tolerance by automatically redistributing the work load assigned to the failed processor(s). This redistribution of load is carried out in a user transparent manner. The effectiveness and generality of the Distributed Genetic Algorithm implementation is demonstrated by solving several problems such as network topology design, network expansion and file allocation.
Get full access to this article
View all access options for this article.