The Energy Saving Management Strategy Of Cloud Data Center Based On Queuing Theory

With the rapid development of Internet and information technology,cloud computing plays an important role in more and more industries.The scale of cloud data center becomes more and more large,which makes the data center produce huge energy consumption,and the energy consumption problem becomes increasingly prominent.In view of its NP complexity and environmental diversity,resource scheduling issues that affect energy consumption have not been well resolved,especially in the context of IT energy consumption becoming a global energy and environmental factor.How to balance energy consumption and performance in the process of resource scheduling is currently a major issue for researchers.Aiming at the problems of high energy consumption and low efficiency in data centers,through analysis and research on the energy saving strategies of existing cloud data centers,the research results obtained in this thesis mainly include:1.Aiming at the problems of task demand diversity and low efficiency in existing task scheduling algorithms,this thesis proposes a density-based clustering strategy for task grouping.Firstly,the strategy uses the K-means algorithm to find the clustering center of the cluster.Secondly,it uses the clustering algorithm proposed in this thesis to group the tasks and narrow the assignment of tasks and resource nodes to similar types of sets.Then,by scheduling similar tasks and resources,the execution efficiency of the clustering algorithm is verified.Finally,the effectiveness of the task grouping clustering strategy is verified through comparative experiments.2.By analyzing the rules of user request task arrival in a cloud environment,this thesis designs an energy-saving queuing task assignment strategy.Firstly,the M/M/n queuing model is used to establish a cloud resource queuing model for the cloud computing system,and the performance of the model is quantitatively analyzed.And a task scheduling model for the cloud computing system is designed.Then,based on this model,the improved particle swarm algorithm is used to complete the scheduling of tasks to resource nodes.Finally,a comparative experiment verifies the practicability and effectiveness of the model in reducing task execution time and energy consumption.Finally,this thesis uses the Spring MVC framework and Java language to design and implement a task distribution system.The two strategies proposed in this thesis areexplicitly verified,and a new solution is proposed for the resource scheduling strategy of the cloud data center.