Research On Virtual Resource Scheduling Strategy In Cloud Computing Environment

Cloud computing technology has changed the traditional computing model and combined with Internet to provided dynamic applications for users. As the core technology of cloud computing, virtualization integrates heterogeneous hardware resources to build virtual resource pool, and use virtual resource scheduling strategy to dynamic deploy and on-demand scheduling the resource pool. Therefore, how to optimize virtual resource scheduling strategy is of great significance in terms of improving the cloud data center resource utilization, reducing resource leak and energy consumption.This paper mainly studies the strategy of virtual resource scheduling in cloud computing environment and the main research contents are as follows:(1) To improve resource utilization and reduce resource leak, this paper proposes VMSM-RB, a virtual machine scheduling model for resource utilization balance in cloud data center, the model abstract the virtual machine scheduling problem as indefinite dimensional vector packing model, and propose an Self-adaptive Particle Swarm Optimization(SA-PSO) to solve the model. Experiments show that virtual resources can be scheduled efficiently by the model and algorithm.(2) To reduce energy consumption of cloud data center, this paper propose VRSA-EO, a virtual resource scheduling algorithm for energy consumption optimization, which considering the energy consumption and the performance of the application. VRSA-EO is divided into two stages: virtual machine initial placement and virtual machine migration. In different stages there are different methods to solve. Experiments show that compared with other virtual resource scheduling algorithms, our algorithm can reducing energy consumption effectively while reducing the virtual machine migration costs.(3) Extending the cloud simulation platform Cloud Sim and implement the virtual resource simulate scheduling platform. Experiments are conducted on the platform to verify virtual machine scheduling model for resource utilization balance VMSM-RB and virtual resource scheduling algorithm for energy consumption optimization VRSA-EO. Experimental results show that the model and algorithm have advantages of improving resource utilization, avoiding resource leaks and reducing energy consumption compared with other classical algorithms.