Joint Route Selection And Rule Cache Algorithm In Software-defined Networking

Software-defined networking(SDN)provides more flexibility and programmability for the network by separating control plane from data plane.To achieve effective data forwarding between any two non-adjacent switches in SDN and ensure network transmission performance,an effective routing and rule caching strategy should be designed.In view of this,according to the characteristics of SDN,this thesis explores joint routing and rule caching algorithms for multi-user flows in SDN.The main research contentsare as follows:The basic concepts and system architecture of SDN are introduced in detail,and the key technologies of SDN are briefly described,then the route selection and rule caching related algorithms are summarized in detail.In addition,the drawbacks of the above algorithms are briefly analyzed.Considering the multi-user flow scenario in SDN network,this thesis studies the joint routing and rule caching problems in SDN under the constraints such as flow conservation and network available resources,etc.In order to emphasize the end-to-end delay performance of the user flows,this thesis formulates the joint routing and rule cache optimization problem as the total end-to-end delay minimization problem.Because the original optimization problem is a non-deterministic polynomial-hard(NP-hard)problem that is difficult to directly solve,it is equivalently converted into routing sub-problem,rule caching sub-problem,and resource sharing sub-problem.The K-shortest path algorithm,the priority-based heuristic algorithm,and the Lagrangian dual method are used to solve the subproblems respectively to obtain joint routing and rule caching optimization strategies.Simulations show the effectiveness of the proposed algorithm.Stressing the scenario where multiple user flow requests dynamically arrive and leave in the network,this thesis proposes an SDN joint routing and rule caching algorithm by applying cost function optimization technique.Jointly considering cache overhead of switches,overhead of control plane and energy consumption due to link transmission,network cost function is defined,and the optimization constraints such as network resource capacity and user flow requirements are examined.Then,the joint routing and rule caching problem is formulated as a cost function minimization problemand a time window-based batch processing scheme is proposed.For the user flows in a specific time window,the priority order is defined,and an improved Dijkstra algorithm and greedy algorithm is presented to solve the problem and to determine the joint routing and rule caching strategy.Simulation results show that the proposed algorithm offers better system performance than previous schemes.