| With the rapid development of the Internet that continues to exceed expectations,a single centralized controller has been unable to meet the needs of today's operators,Internet enterprises and users.It is estimated that the future output value of SDN will reach$70.41 billion in 2024.It is an inevitable trend that the single controller architecture will be extended to the multi-controller architecture.The practical Byzantine algorithm is one of the multi-controller synchronization algorithms,but its communication and time complexity reaches O(~2).At the same time,in the process of distributed SDN routing,the switch topology controlled by each controller needs to be considered when calculating the logic node topology at the control level.It is not consistent with the actual situation to select the path only according to the parameters of the controller.The main work of this is as follows:1.A high performance distributed software definition network vice master node Byzantine fault tolerance(HVMBFT)algorithm is designed for distributed software-defined networks.By optimizing the synchronization process and adding a trusted mechanism,In the case that the malicious node acts as the master node and does evil,it is proposed to keep the current master node.Firstly,the secondary master node is selected through the secondary master node algorithm and the synchronization data of the master node is carried out.If the newly selected secondary master node does evil,the secondary master node algorithm will be run again.Since it is specified that there will only be f malicious nodes in the system,the election will only occur f-1 times at most.Moreover,due to the existence of credibility,the secondary master node will not be malicious nodes next time.In the best case,the secondary master node algorithm can reduce the communication complexity and time complexity required for synchronization to O(),and in the worst case,it can only occur once as O(~2).The simulation results show that when the number of nodes is more than or equal to seven nodes,the lowest synchronization time reduction rate and traffic reduction rate of the algorithm are 7.47%and 41.67%,respectively,and the highest are 55.03%and 66.67%,respectively.The improved algorithm is applied to the distributed SDN synchronization system,and the system performs well.2.An improved ant colony algorithm is designed based on the number of all feasible end-to-end links within the control node as one of the heuristic factor parameters.In this algorithm,the number of all feasible end-to-end links,the average bandwidth of the control node and the load of the control node are taken as the parameters to calculate the heuristic factor.Based on the state transition equation of ACS,the maximum node is removed from the optional list of randomly selected modes as a new state transition equation,and on this basis,a new pheromone updating method is proposed.The path obtained by the ant colony algorithm is synchronized to each node through the synchronization algorithm,and each node calculates its own internal path and issues a flow table.The simulation results show that the bandwidth processing ability of the proposed algorithm is 20.48%higher than that of the optimal algorithm in the comparison algorithm. |
PDF Full Text Request