Research On Virtual Machine Scheduling Strategy Of Cloud Computing

Virtual machine scheduling is a key technology of cloud computing resource management. In cloud computing resource management, the number and load of the virtual machines change frequently with the demand of the users and applications. Static resource allocation tends to make the virtual machine produces waste or shortage of resources, therefore the virtual machine scheduling directly affects the overall system resources utilization, energy consumption and quality of service. At present, most of cloud computing data centers which are established by the IT companies take simple scheduling strategy to achieve the virtual machine scheduling. With the continuous expansion of the scale of cloud computing, virtual machine scheduling problem of the cloud computing environment becomes more and more complicated, the traditional static and manual scheduling methods have been difficult to meet the increasingly stringent requirements of cloud management platform.In order to solve virtual machine scheduling problem of the cloud computing more effectively, an optimized virtual machine scheduling strategy is proposed in this paper. According to the monitoring of the resources utilization, we classify the physical hosts in the cloud data center. Aiming at the high load and low load physical hosts, scheduling approach is proposed based on the technology of virtual machine migration, which prevents that the physical host resource utilization is too high to reduce physical host’s availability as well as the physical host resource utilization is too low to waste of resources. Aiming at the physical hosts in the normal state, the virtual machine dynamic scheduling method based on the characteristics of flexibility is presented, the method allocates resources for the virtual machine of physical host dynamically, so that the performance of virtual machine could keep stable and the virtual machine could provide a more efficient service. Finally, the simulation experiment results show that the strategy has some advantages which could reduce energy consumption of cloud data center and save network flow.