Abstract:

In consideration of low efficiency, resource load imbalance of the traditional task scheduling algorithm and other shortcomings, and the heterogeneous and dynamic characteristics of a modern grid system, an effective adaptive update method for the crossover and mutation probability was proposed based on the genetic algorithms, which improves the global search ability and convergence speed of the genetic algorithm. Simulation results show that improved genetic algorithm enhances the speed of evolution significantly, can better handle grid task scheduling, improve task scheduling efficiency and reduce resource load imbalance with high practicality.

Key words: genetic algorithm, adaptive, grid, task scheduling