Genetic algorithms (GAs) are powerful search techniques that are used successfully to solve problems in many diierent disciplines. Parallel GAs are particularly easy to implement and promise substantial gains in performance and as such there has been extensive research in this eld. This paper attempts to collect, organize, and present in a uniied way some of the most representative publications...

In this work we review the most important existing developments and future trends in the class of Parallel Genetic Algorithms (PGAs). PGAs are mainly subdivided into coarse and fine grain PGAs, the coarse grain models being the most popular ones. An exceptional characteristic of PGAs is that they are not just the parallel version of a sequential algorithm intended to provide speed gains. Instea...

The availability of massively parallel computers makes it possible to apply genetic algorithms to large populations and very complex applications. Among these applications are studies of natural evolution in the emerging eld of articial life, which place special demands on the genetic algorithm. In this paper, we characterize the di erence between panmictic and local selection/mating schemes in...

This paper extends previous analyses of parallel GAs with multiple populations (demes) to consider cases where the demes are connected in regular topologies with a xed number of neighbors. The calculations in this paper make explicit the relations between the deme size, the migration rate, and the degree of the connectivity graph with the probability that the desired solution will be found. An ...

In this paper we investigate the design of highly parallel Genetic Algorithms. The Traveling Salesman Problem is used as a case study to evaluate and compare diierent implementations. To x the various parameters of Genetic Algorithms to the case study considered, the Holland sequential Genetic Algorithm, which adopts diierent population replacement methods and crossover operators, has been impl...

Parallel computers offer massive potential in the domain of machine learning for investigating domains with domain theories more complex than those tractable with classical inductive algorithms. However to exploit the power of parallel computers it is necessary to develop an algorithm that works more efficiently in parallel than in serial. This paper outlines one approach to implementing a para...

Genetic algorithms (GAs) are powerful search techniques that are used successfully to solve problems in many different disciplines. Parallel GAs are particularly easy to implement and promise substantial gains in performance. As such, there has been extensive research in this field. This survey attempts to collect, organize, and present in a unified way some of the most representative publicati...

Parallel genetic algorithms are often very different from the \traditional" genetic algorithm proposed by Holland, especially with regards to population structure and selection mechanisms. In this paper we compare several parallel genetic algorithms across a wide range of optimization functions in an attempt to determine whether these changes have positive or negative impact on their problemsol...

This paper examines the scalability of several types of parallel genetic algorithms (GAs). The objective is to determine the optimal number of processors that can be used by each type to minimize the execution time. The first part of the paper considers algorithms with a single population. The investigation focuses on an implementation where the population is distributed to several processors, ...

This paper describes a parallel genetic algorithms library which enables easy development of parallel programs exploiting genetic algorithm approach. The library is based on TOLKIEN C++ sequential library. SPMD paradigm is applied and PARA++ library is used for message passing. Island and global population models of parallelism have been implemented. The library may be used on virtual network c...