【摘要】 冗余资源是提升网络可靠性的关键元素，针对软件定义网络(SDN)数据平面优化设计过程中存在冗余链路难以合理分配的问题，首先改进了SDN数据平面可靠性评估方法，能够准确评估冗余链路分配方案对于SDN整体可靠传输能力的影响；然后构建了以满足冗余链路成本约束、最大化可靠传输能力为目标的SDN可靠性加固模型；最后提出了一种基于改进遗传算法的SDN可靠性加固算法SDNRH-IGA来求解该模型。对比实验结果表明，文中提出的算法能够搜寻到全局最优冗余链路分配方案，且搜寻到最优解的概率和速度均优于同类算法。更多还原

【Abstract】 Redundant resource is the key element to improve the network reliability. To overcome the difficulty in allocating the redundant link resources reasonably in the process of SDN data plane optimization design. Firstly, the reliability evaluation method of SDN data plane is improved to accurately measure the impact of redundant link allocation scheme on the overall reliable transmission capacity of the network. Then, a reliable strengthening model of SDN is designed to satisfy the cost constraints of redundant links and maximize the reliable transmission capacity for the allocation of redundant links. Finally, a reliability strengthening algorithm SDNRH-IGA for SDN based on improved genetic algorithm is proposed to solve the model. Experiment results show that the proposed algorithm can find the global optimal redundant link allocation scheme, and the probability and speed of finding the optimal solution are better than other algorithms.更多还原