On the feasibility and efficacy of control traffic protection in software-defined networks

• Published in: Science China. Information sciences Vol. 58; no. 12; pp. 1 - 19
• Main Authors: Hu, YanNan, Wang, WenDong, Gong, XiangYang, Que, XiRong, Cheng, ShiDuan
• Format: Journal Article
• Language: English
• Published: Beijing Science China Press 01.12.2015; Springer Nature B.V
• Subjects: Algorithms; Computer Science; Controllers; Information Systems and Communication Service; Network latency; Networks; Performance degradation; Planes; Research Paper; Software; Software-defined networking; Switches; Switching theory; Topology; Traffic control; Traffic engineering; Traffic flow; 控制流; 120104; optimization; 优化; 保护性; 软件定义网络; protection; software-defined network; control traffic; resilience; 容错性
• ISSN: 1674-733X; 1869-1919
• DOI: 10.1007/s11432-015-5483-7

Software Defined Networking (SDN) is an emerging networking paradigm that assumes a logically centralized control plane separated from the data plane. Despite all its advantages, separating the control and data planes introduces new challenges regarding resilient communications between the two. That is, disconnections between switches and their controllers could result in substantial packet loss and performance degradation. This paper addresses this challenge by studying the issue of control traffic protection in SDNs with arbitrary numbers of controllers. Specifically, we propose a control traffic protection scheme that combines both local rerouting and constrained reverse path forwarding protections, through which switches can locally react to failures and redirect the control traffic using standby backup forwarding options. Our goal is then to find a set of primary routes for control traffic, called protected control network, where as many switches as possible can benefit from the proposed protection scheme. We formulate the protected control network problem, prove its NP-hardness, and develop an algorithm that reconciles protectability and performance (e.g., switch-to-control latency). Through extensive simulations based on real topologies, we show that our approach significantly improves protectability of control traffic. The results should help further the process of deploying SDN in real-world networks.