Research On Node Deployment And Energy Saving Strategy Based On Hybrid SDN Network

The emergence of Software Defined Networking(SDN)has added new impetus to the innovation and development of the network,and it has brought broad development and prospects to the network.In response to the rapid increase in Internet power consumption caused by the continuous expansion of the Internet in recent years,energy conservation has become an important issue in network research.The flexible centralized control features of SDN networks have brought excellent opportunities to improve network performance and reduce energy consumption.However,the transition from traditional networks to SDN networks requires significant changes to existing network architectures and equipment.Due to limitations in upgrade budgets and technical issues,it is necessary to gradually replace the limited traditional switching equipment in the backbone network with SDN equipment.The mode of coexistence of traditional equipment and SDN equipment will exist for a long time,so the research on node replacement and energy saving mechanism for hybrid SDN networks is worth exploring.In view of the lack of research schemes on deployment strategies of SDN nodes in the current hybrid SDN network,this thesis first proposes a topology-based hybrid SDN network node deployment scheme.Based on the node and link information of the network topology,and within the constraints of the node's deployment budget,the solution will find a traditional node set for the network that needs to be replaced with SDN node set,so it can maximize the number of controllable links in the network.At the same time,considering the demand of a hybrid network that divides itself into multiple areas for management,this thesis also proposes a node deployment scheme that meets the requirements of multiple areas in a hybrid SDN network under the constraints of the deployment budget.For some scenarios where the traffic matrix in the network can be measured,this thesis also proposes to choose another node deployment scheme,that a node set that meets budget constraints and can maximize the network flow control capability.Combined with the SDN switching equipment replacement scheme proposed in this thesis,in order to study about the energy-saving problem of hybrid SDN networks,this thesis proposes an energy-saving algorithm based on genetic algorithm.Under the condition of ensuring the normal operation of the network,the state of some controllable links is adjusted to the sleeping state,so the redundant load of the network is reduced and the algorithm can achieve the energy-saving goal of the hybrid SDN network.At the same time,in order to meet the demands of obtaining energy-saving settings solutions quickly in the network,this thesis also proposes a fast link control algorithm that uses greedy ideas.Under the condition that the traffic and link load constraints in the network are satisfied,the link with a lower load in the current SDN link set will be preferentially selected to adjust its state.Through a series of experiments using real traffic matrix information in the ISP topology and comparison with other related work,these experiments prove that the combination of node deployment and energy-saving algorithms proposed in this paper can effectively reduce network energy consumption.