基于粒子群算法和RBF神经网络的云计算资源调度方法研究

为了获得云计算资源调度的多目标优化方案,提出了一种云计算资源的动态调度管理框架；然后给出了本系统的基本架构形式,并对其进行了详细设计；其次,建立了以提高应用性能、保证云应用的服务质量和提高资源利用率为目标的多目标优化模型,并结合最新的RBF神经网络和改进粒子群算法对其求解；最后,在CloudSim平台进行了仿真,实验结果表明提出的框架及算法能有效减少虚拟机迁移次数和物理结点的使用数量,在提高资源利用率的同时,能保证云应用的服务质量。

Research on Resources Scheduling Method in Cloud Computing Based on PSO and RBF Neural Network

In order to implement the multi-objective optimization scheme in cloud computing system,firstly,a dynamic management framework was proposed,providing the structure of the resources scheduling in cloud computing system.Secondly,a multi-objective optimization model was established,which ensures the quality of cloud applications and improves the utilization rate of resources.The RBF neural network and improved particle swarm algorithm were combined to solve the model.Finally, the result of the experiment on the CloudSim simulation platform indicates that the framework and the proposed algorithm can effectively reduce the number of virtual machine migration and the number of used physical nodes,and the scheduling system can not only improve the utilization rate of resources,but also ensure the QoS of cloud application.