Design And Implementation Of Energy Efficient Routing Mechanisms In Data Center Networks

To satisfy the increasing users' requirements,data centers are growing exponentially.However,the development of data centers brings in the energy consumption problem.The energy consunption of data centers is mainly generated from three aspects,i.e.,network infrastructures,servers,and cooling systems.As the servers and cooling systems are becoming more energy efficient,it is expected that the power consumption share of the networking devices within a data center will increase.It has been projected that the energy consumption of the network component will increase by up to 50%within data centers.In addition,DCNs(Data Center Networks)provide multiple paths for all the end hosts.But redundant transmission paths make the bandwidth resource graveness waste.Existing statistics show that the average link utilization of DCNs is around 5%?25%.It also aggravates the energy consumption problem.Therefore,we must introduce techniques to solve or alleviate the high energy consumption of DCNs.Two routing mechanisms,i.e.,routing mechanism based on MBOA(Migrating Birds Optimization Algorithm)and routing mechanism based on FOA(Fruit fly Optimization Algorithm)are proposed to solve the problem of high energy consumption in DCNs.At first,according to the characteristics of DCNs,the request situation of business flows,the positive correlation relationship between the switching equipment's power and their load,the dynamical power adjustment strategy,this thesis deduces a method calculating the switching equipment's power in DCNs.Then,to reflect the energy consumption situation in DCNs more clearly,this thesis designs the energy efficient indicator,which is used to represent the business flows that unit energy consumption can carry.At last,two routing mechanisms are designed respectively.The MBOA-based routing mechanism chooses the best energy efficient routing paths for business flows by regarding each path as a bird and simulating the behavior of selecting a leading bird from migratory birds.The FOA-based routing mechanism chooses the best energy efficient routing paths for business flows by simulating the strategy of fruit flies flying to the comestible when they are looking for food.This thesis simulates the proposed routing mechanisms,and compares them to the benchmark algorithm from the perspective of energy efficiency,blocking rate and load balancing rate.Simulation results reveal that the routing mechanisms designed in this thesis are superior to algorithm compared.Therefore,the mechanisms proposed in this thesis are feasible and effective.