Predicting Speedups of Idealized Bounding Cases of Parallel Genetic Algorithms

This paper presents models that predict the speedup of two cases that bound the possible topologies and migration rates of parallel genetic algorithms (GAs). The rst bounding case is a parallel GA with completely isolated demes or subpopulations and for this case the model and the experiments show that the speedup is not very signiicant when more demes are used. The second model predicts the speedup when each deme communicates with every other deme using a maximal migration rate. For this case, we show that when the communication time is not constant there is a combination of number of demes and deme size that maximizes the speedup. The models are validated with computational experiments using functions of varying dii-culty.