| 控制器故障下的软件定义机载网络数据平面迁移策略 |
| |
| 引用本文： | 吕娜,潘武,陈柯帆,张彦晖. 控制器故障下的软件定义机载网络数据平面迁移策略[J]. 航空学报, 2021, 42(3): 324228-324228. DOI: 10.7527/S1000-6893.2020.24228 |
| |
| 作者姓名： | 吕娜 潘武 陈柯帆 张彦晖 |
| |
| 作者单位： | 空军工程大学信息与导航学院,西安710077;中国人民解放军94860部队,南京210000 |
| |
| 基金项目： | 国家自然科学基金（61703427） |
| |
| 摘 要： | 软件定义网络（SDN）为突破机载网络功能僵化所造成的瓶颈提供了新的契机。针对软件定义机载网络控制域内控制节点故障问题，提出一种基于时延和负载均衡的传输节点迁移策略，将故障控制节点所管控数据平面中的传输节点迁移至其他正常控制节点，恢复网络的正常管控。以迁移时延和负载均衡率为目标，建立迁移优化模型，提出基于时延和负载均衡的传输节点迁移算法，指导数据平面中的迁移动作，最终对两个性能目标进行综合权衡。实验结果表明，与距离就近迁移（DCM）策略和分布式逐跳（DHA）策略相比，所提策略在迁移时延和负载均衡率2个方面进行了合适的折中，避免出现迁移时延过大或控制节点负载失衡的情况，有效解决机载网络故障恢复问题。
|
| 关 键 词： | 机载网络 软件定义网络(SDN) 数据平面 迁移时延 负载均衡率 |
| 收稿时间： | 2020-05-15 |
| 修稿时间： | 2020-06-06 |
Data plane migration strategy of software-defined airborne network with controller failure |
| |
| LV Na,PAN Wu,CHEN Kefan,ZHANG Yanhui. Data plane migration strategy of software-defined airborne network with controller failure[J]. Acta Aeronautica et Astronautica Sinica, 2021, 42(3): 324228-324228. DOI: 10.7527/S1000-6893.2020.24228 |
| |
| Authors: | LV Na PAN Wu CHEN Kefan ZHANG Yanhui |
| |
| Affiliation: | 1. College of Information and Navigation, Air Force Engineering University, Xi'an 710077, China;2. Unit 94860 of the PLA, Nanjing 210000, China |
| |
| Abstract: | Software-Defined Networking (SDN) provides a new opportunity to break the bottleneck caused by the rigidity of airborne network functions. Aiming at the problem of control node failure in the software-defined airborne network control domain, we propose a transmission node migration strategy based on time delay and load balancing to transfer the transmission nodes in data plane that are controlled by the failed control node to other normal control nodes to restore the normal control of the network. With the migration delay and load balancing rate as the goal, a migration optimization model is established, and the transmission node migration algorithm based on delay and load balancing is proposed to guide the migration action in the data plane, and finally the two performance goals are comprehensively balanced. The experimental results show that, compared with the Distance Closest Migration (DCM) strategy and the Distributed Hopping Algorithm (DHA) strategy, the proposed strategy makes a suitable compromise between the migration delay and the load balancing rate, avoids excessive migration delay or unbalanced control node load, and effectively solves the problem of airborne network fault recovery. |
| |
| Keywords: | airborne networks Software-Defined Networks(SDN) data plane migration delay load balancing rate |
| 本文献已被 万方数据 等数据库收录！ |
| 点击此处可从《航空学报》浏览原始摘要信息 |
|
点击此处可从《航空学报》下载全文 |
|
