306 IEEE TRANSACTIONS ON NETWORK AND SERVICE MANAGEMENT, VOL. 9, NO. 3, SEPTEMBER 2012 A Unified Approach to Routing Protection in IP Networks Qi Li, Member, IEEE, Mingwei Xu, Member, IEEE, Jianping Wu, Fellow, IEEE, Patrick P. C. Lee, Member, IEEE, Xingang Shi, Member, IEEE, Dah Ming Chiu, Fellow, IEEE, and Yuan Yang, Student Member, IEEE Abstract—Routing failures are common on the Internet and routing protocols can not always react fast enough to recover from them, which usually cause packet delivery failures. To address the problem, fast reroute solutions have been proposed to guarantee reroute path availability and to avoid high packet loss after network failures. However, existing solutions are often specific to single type of routing protocol. It is hard to deploy these solutions together to protect Internet routing including both intra- and inter-domain routing protocols because of their individual computational and storage complexity. Moreover, most of them can not provide effective protection for traffic over failed links, especially for the bi-directional traffic. In this paper, we propose a unified fast reroute solution for routing protection under network failures. Our solution leverages identifier based direct forwarding to guarantee the effectiveness of routing protection and supports incremental deployment. In particular, enhanced protection cycle (e-cycle) is proposed to construct rerouting paths and to provide node and link protection for both intra- and inter-domain routing protocols. We evaluate our solution by simulations, and the results show that the so- lution provides 100% failure coverage for all end-to-end routing paths with approximately two extra Forwarding Information Base (FIB) entries. Furthermore, we report an experimental evaluation of the proposed solution in operational networks. Our results show that the proposed solution effective provides failure recovery and does not introduce processing overhead to packet forwarding. Index Terms—IP networks; routing; routing protection; re- silience. Manuscript received April 9, 2011; revised October 5, 2011, January 25 and March 28, 2012; accepted April 6, 2012. The associate editor coordinating the review of this manuscript and approving it for publication was J. Sch¨ onw¨ alder. This work was supported by the National Natural Science Foundation of China under Grant No. 61073166, Grant No. 61133015, and Grant No. 61161140454; by the National Basic Research Program of China (973 Program) under Grant No. 2009CB320502 and Grant No. 2012CB315803; and by the National High-Tech Research and Development Program of China (863 Program) under Grant No. 2011AA01A101. The preliminary version of this paper titled “Achieving Unified Protection for IP Routing,” was published in the Proceedings of the 19th International Conference on Computer Com- munications and Networks (ICCCN), 2010 [28]. This journal version makes the following extensions to the conference paper: (i) we specifically address BGP (iBGP/eBGP) node protection, which is not addressed in the literature; (ii) we evaluate the performance of different existing protection solutions in the GEANT and Rocketfuel networks, and compare the overhead of these solutions; (iii) we discussed the implications for real deployment of e-cycle, and demonstrate the practicality of e-cycle by deploying it in operational networks. Q. Li, M. Xu, J. Wu, and Y. Yang are with the Department of Computer Science, Tsinghua University, Beijing, China, 100084 (e-mail: {liqi, xmw, jianping, yyang}@csnet1.cs.tsinghua.edu.cn). P. Lee is with the Department of Computer Science and Engi- neering, The Chinese University of Hong Kong, Hong Kong (e-mail: pclee@cse.cuhk.edu.hk). X. Shi is with the Network Research Center, Tsinghua University, Beijing, China, 100084 (e-mail: shixg@cernet.edu.cn). D. Chiu is with the Department of Information Engineering, The Chinese University of Hong Kong, Hong Kong (e-mail: dmchiu@ie.cuhk.edu.hk). Digital Object Identifier 10.1109/TNSM.2012.070512.110138 I. I NTRODUCTION I NTERNET routing connects different IP networks and plays a critical role in ensuring packet delivery throughout the Internet. However, previous studies show that current routing systems are ineffective in recovering from routing failures. For instance, the 2006 earthquake in Taiwan caused global disruption in the Internet although there remained unaffected links that could provide potential connectivity for different IP networks. Even though most routing failures, such as Border Gateway Protocol (BGP) session resets and transient hardware failures, are short-term and last less than 3 minutes [1], current routing protocols fail to react quickly to recover from such short-term routing failures. It is not unusual for routing protocols to take several minutes or even longer to converge [2]. This significant recovery time leads to unreliable packet delivery. Extensive research has been conducted in order to effec- tively address routing failures. One line of work is to develop solutions that provide fast routing convergence, which has been extensively studied in the literature [3], [4]. However, none of these solutions has been deployed in operational networks due to their complexity, or in some cases, due to subtle design flaws. For instance, Ghost Flushing [3] expedites convergence by sending extra route withdrawal messages but may exacerbate routing convergence in failover events. Basically, fast routing convergence is not effective for handling routing blackholes and loops. Another line of work that addresses routing failures is to realize routing protection by using backup routing paths [5], [6], [7], [8], [9], [10], i.e., fast reroute approaches. However, such approaches again have different design limitations. IP- FRR solutions [6], [7], which are active subjects in the IETF, focus only on the protection of intra-domain routing. They share important drawbacks such as difficult deployment and/or uncertain protection effectiveness over failed links [11]. To support fast reroute in inter-domain routing, Bonaventure et al. [5] propose BGP fast reroute (BGP-FRR), the first solution that protects external BGP (eBGP) between different ASes by automatic protection and aims to realize effective protection by extra manual configurations. R-BGP [12] is proposed to provide automatic failover for eBGP failures. R-BGP requires an extra Forwarding Information Base (FIB) entry for every prefix under protection, and the protection effectiveness is greatly restricted by routing policies. Also, neither BGP-FRR nor R-BGP considers internal BGP (iBGP) failures [13], [14]. Furthermore, previous studies consider protection only for a single type of routing protocol, either intra- or inter-domain routing. It would be a difficult task to deploy these solutions 1932-4537/12/$31.00 c  2012 IEEE