Research On The Virtual Machines Scheduling Energy Saving Method Of Cloud Platform Based On Improved PSO Algorithm And The Double Layer Management Model

With the rapid development of cloud computing platform, the problem of energy consumption is becoming more and more prominent. The main idea of energy saving method in in the cloud platform which is used virtualization technology is to integrate the virtual machine on some physical servers, and turns off the redundant server to achieve the goal of energy saving.When reseaching the energy conservation in the cloud platform which is used virtualization technology,we find there is a single node to manage the whole platform and be responsible for all virtual machine migration and schedule in the existing model.This leads to the problem of large scale and low efficiency of migration.But if the management node crash, management function of platform will be failure. Therefore, according to the above description,the following research in this paper:Firstly, a double layer management model of virtual machine is proposed to solve the problem of single node bottleneck and largescope of virtual machine migration. Greatly reducing the management pressure of each management node, enhancing the reliability of the platform.At the same time reducing the size of the virtual machine migration problem. And speeding up find virtual machine deployment scheme.Second, an improved PSO algorithm is proposed to calculate the virtual machine migration scheme. The virtual machine migration problem is a multi-objective optimization problem. The application of PSO algorithm in discrete multi objective optimization problem is improved and reselect the optimization objective which combined with the factors of energy consumption to achieve energy saving and optimization under the premise of ensuring the service performance.Third, after validating on the CloudSim simulation platform, the improved PSO algorithm proposed in this paper have obvious energy-saving effect in cloud platform.