A Task Scheduling Algorithm Based on Genetic Algorithm and Ant Colony Optimization in Cloud Computing
Chunyan Liu, Chengming Zou, Pei Wu · 2014
An efficient approach to task scheduling algorithm remains a long-standing challenge in cloud computing. In spite of the various scheduling algorithms proposed for cloud environment, those are mostly improvements based on one algorithm. And it's easy to overlook limitations of the algorithm itself. Aiming at characteristics of task scheduling in cloud environment, this paper proposes a task scheduling algorithm based on genetic-ant colony algorithm. We take the advantage of strong positive feedback of ant colony optimization (ACO) on convergence rate of the algorithm into account. But the choice of the initial pheromone has a crucial impact on the convergence rate. The algorithm makes use of the global search ability of genetic algorithm to solve the optimal solution quickly, and then converts it into the initial pheromone of ACO. The simulation experiments show that under the same conditions, this algorithm overweighs genetic algorithm and ACO, even has efficiency advantage in large-scale environments. It is an efficient task scheduling algorithm in the cloud computing environment.