Research On Data Center Energy Saving Optimization Based On Virtualization And Dynamic Environment Resource Coordination Scheduling

In the era of rapid development of information technology,data center,as the carrier of information,is expanding constantly,which brings great impetus to social development,and at the same time,it also causes serious energy waste.At present,there are many researches on energy saving optimization of data centers,but they are usually aimed at energy saving measures for single devices or modules.The data center lacks coordination,resulting in poor energy saving effect.Therefore,on the basis of fully studying the data center structure and energy consumption factors,the whole data center is divided into IT resources and dynamic environment resources,and a mutually synergetic energy-saving framework is put forward.In terms of IT resource optimization,we firstly analyze the existing research results,grasp the relevant principles of virtualization,and build the server energy consumption model.Because the allocation and migration of virtual machines on servers is very important for energy saving optimization,in this paper,a virtual machine allocation algorithm based on genetic algorithm and cat swarm optimization is proposed,which makes the allocation of virtual machines on the server optimal.The first order adaptation algorithm is used to migrate the virtual machine on the server,reduce the number of idle servers to avoid unnecessary energy consumption,and then use CloudSim to simulate the proposed method.In the aspect of energy saving and optimization of dynamic environment resources,by analyzing the influence factors of data center temperature distribution,a prediction method of temperature distribution based on machine learning technology is put forward on the basis of the correlation between server and air conditioning system.Then according to the server thermal load transfer effect on temperature field of developed control strategy of air conditioning system,Keihanna data center at Gyeonggi Do area in Japan established as a case study,the relevant theory and the proposed method was verified,and the influence factors and model performance results are discussed.Through the optimization of the distribution and migration of virtual machine to achieve energy efficiency optimization of the server,and then according to the scheduling task server generated by the thermal load effect on the temperature field,using machine learning to predict the temperature distribution of the data center,in order to build a mutually synergetic energy-saving framework,finally realize the overall energy efficiency of data center.