A Revenue-Driven Adaptive Heterogeneous Task Distribution Method for Cloud Computing

In the vigorous development era of Internet-based applications, the demand for cloud computing is getting higher and higher. To solve the task scheduling in cloud computing, we propose a task scheduling model based on genetic algorithm. Gaussian mutation is to improve traditional genetic algorithms to enhance the adaption to the change of our scheduling model. Quick sort and optimal selection introduce to improve fitness function, making screening faster and more convenient. The crossover and mutation operations with rule constraints are use to improve the quality of individuals. The results of simulation experiments show that the improved algorithm can solve the task scheduling problem in cloud computing environment more effectively.