Font Size: a A A

Energy-saving And Traffic Engineering In Hybrid SDN

Posted on:2019-03-20Degree:MasterType:Thesis
Country:ChinaCandidate:S ChengFull Text:PDF
GTID:2428330566960641Subject:Computer Science and Technology
Abstract/Summary:PDF Full Text Request
The emergence of software defined networking has injected new vitality into the development of the network.However,due to numerous constraints such as economy,technology,and talent,operators and researchers cannot directly evolve traditional networks into SDN networks,leading to many solutions that cannot be realized in the short term.In order to resolve these inconsistencies,the hybrid SDN network adds a centralized control system by introducing SDN switches in a conventional network,so that the controller can indirectly grasp the network operating status,monitor network traffic,and execute optimization policies through these SDN switches.The hybrid SDN network can reduce the deployment cost of the SDN network.However,for the traffic scheduling and optimization of the hybrid SDN network,there is still a lack of relevant research on reducing link energy consumption and resource conservation.In addition,in the study of the optimization strategy of the hybrid SDN network,there is a lack of a research scheme for the SDN switch deployment strategy.For energy saving and traffic optimization in hybrid SDN networks,this paper designs and implements an SDN switch deployment location selection and link behavior optimization scheme(EaT hSDN,Energy saving And Traffic engineering in hybrid SDN networks).This solution reduces link energy consumption by reducing the link redundancy load.Under the premise of guaranteeing correct packet forwarding and limited maximum utilization of the link,the scheme is based on improved genetic algorithm and heuristic algorithm.By optimizing the deployment position of the SDN switch,defining the separation ratio of the switch to the traffic,closing some invalid links,and adjusting the maximum load of the link,the energy consumption of the link is reduced.At the same time,based on some public modules of EaT hSDN,this paper also implements a load balancing algorithm that enforces the controllable flow of all hybrid SDN networks.In order to verify the effectiveness and accuracy of the scheme,this paper tests the performance of EaT h SDN by building a simulation platform,and compares it horizontally with related existing algorithms.The experimental results show that under the premise of only 20% of SDN switch deployment budget,EaT hSDN can reduce link energy consumption by more than 30%.
Keywords/Search Tags:Software Defined Networking, Hybrid SDN network, Green Network, Traffic engineering, Load balancing, Genetic algorithm
PDF Full Text Request
Related items