降低数据中心能耗提高网络服务质量优化算法研究-电气工程专业论文.docxVIP

ABSTRACT The energy consumption of networking elements in modern cloud computing data centers accounts for a large part of its total power consumption. With the development of virtual machine technique and programming switch technology, more and more control strategies and optimization methods have been proposed to reduce the energy consumption of networking elements and improve the quality of services of the data center network. In this thesis, I firstly introduce and analyze the virtual machine technology and the characteristics of the communication traffic flows in cloud computing data centers. And then, with regard to the two main services in modern cloud computing data centers, which are common internet business handled by MapReduce and kinds of scientific workflows, I propose “advanced genetic algorithm” and “job-aware VM placement and route scheduling” algorithm (JAVPRS) respectively. Considering the huge amount of virtual machines and physical machines in modern data centers, I propose a novel gene encoding method for the genetic algorithm which is for optimizing the placement of virtual machines carrying MapReuce jobs. The gene encoding method is a two-dimensional array composed by identifier of virtual machines and physical machines. The crossover operation of the genetic algorithm is base on this kind of encoding method, and the cutting point is the middle of the two independent dimensions. This kind of crossover operation not only persist the coupling property of the virtual machines, but also guarantee the solution space traversal. In the JAVPRS algorithm for scientific computing jobs, I migrate the jobs whose communication time is larger than the time threshold (TH) to “neighbour” physical machines according to the time feature of communication data flows, and consolidate the communication data flows into part of the communication links in order to reduce the amount of working switches and balance the communication load in powered on switches and link