Eric Keller
Jennifer Rexford
Abstract
Traditional traffic engineering adapts the routing of traffic within the network to maximize performance. We propose a new approach that also adaptively changes where traffic enters and leaves the network---changing the "traffic matrix", and not just the intradomain routing configuration. Our approach does not affect traffic patterns and BGP routes seen in neighboring networks, unlike conventional inter-domain traffic engineering where changes in BGP policies shift traffic and routes from one edge link to another. Instead, we capitalize on recent innovations in edge-link migration that enable seamless rehoming of an edge link to a different internal router in an ISP backbone network---completely transparent to the router in the neighboring domain. We present an optimization framework for traffic engineering with migration and develop algorithms that determine which edge links should migrate, where they should go, and how often they should move. Our experiments with Internet2 traffic and topology data show that edge-link migration allows the network to carry 18.8% more traffic (at the same level of performance) over optimizing routing alone.
- M. Agrawal, S. Bailey, A. Greenberg, J. Pastor, P. Sebos, S. Seshan, J. van der Merwe, and J. Yates. RouterFarm: Towards a dynamic, manageable network edge. In SIGCOMM Workshop on Internet Network Management, September 2006. Google Scholar
Digital Library
- D. G. Andersen, H. Balakrishnan, M. F. Kaashoek, and R. Morris. Resilient overlay networks. In Proc. ACM SOSP, October 2001. Google ScholarDigital Library
- D. Applegate, L. Breslau, and E. Cohen. Coping with network failures: Routing strategies for optimal demand oblivious restoration. In Proc. ACM SIGCMETRICS, June 2004. Google ScholarDigital Library
- A. Elwalid, C. Jin, S. Low, and I. Widjaja. MATE: MPLS adaptive traffic engineering. In Proc. IEEE INFOCOM, 2001.Google ScholarCross Ref
- B. Fortz and M. Thorup. Internet traffic engineering by optimizing OSPF weights. In Proc. IEEE INFOCOM, 2000.Google ScholarCross Ref
- Internet2. http://www.internet2.org.Google Scholar
- S. Kandula, D. Katabi, B. Davie, and A. Charny. Walking the tightrope: Responsive yet stable traffic engineering. In Proc. SIGCOMM, 2005. Google ScholarDigital Library
- E. Keller, J. Rexford, and J. van der Merwe. Seamless BGP session migration with router grafting. In Proc. Networked Systems Design and Implementation, April 2010. Google ScholarDigital Library
- E. Keller, M. Schapira, and J. Rexford. Rehoming Edge Links for Better Traffic Engineering. Technical Report TR-917-11, Princeton University Computer Science Department, 2011.Google Scholar
- R. Mahajan, D. Wetherall, and T. Anderson. Negotiation-based routing between neighboring ISPs. In Proceedings of the 2nd Symposium on Networked Systems Design and Implementation, Boston, MA, USA, April 2005. Google ScholarDigital Library
- L. Qiu, Y. R. Yang, Y. Zhang, and S. Shenker. Selfish routing in Internet-like environments. In Proc. SIGCOMM, 2003. Google ScholarDigital Library
- M. Roughan and Y. Zhang. GATEway: symbiotic inter-domain traffic engineering'. In The Second International Workshop on Game Theory in Communication Networks, Athens, Greece, October 2008. Google ScholarDigital Library
- S. Savage, T. Anderson, A. Aggarwal, D. Becker, N. Cardwell, A. Collins, E. Hoffman, J. Snell, A. Vahdat, G. Voelker, and J. Zahorjan. Detour: A case for informed Internet routing and transport. IEEE Micro, January 1999. Google ScholarDigital Library
- R. Szabó, A. Takács, and A. Császár. Optimised multi homing - an approach for inter-domain traffic engineering. In Proceedings of the 2nd International Workshop on Inter-Domain Performance and Simulation (IPS2004), Budapest, Hungary, March 2004.Google Scholar
- R. Teixeira, T. Griffin, M. G. C. Resende, and J. Rexford. TIE breaking: Tunable interdomain egress selection. IEEE/ACM Trans. Networking, August 2007. Google ScholarDigital Library
- H. Wang, H. Xie, L. Qiu, Y. R. Yang, Y. Zhang, and A. Greenberg. COPE: Traffic engineering in dynamic networks. In Proc. SIGCOMM, 2006. Google ScholarDigital Library
- C. Zhang, Z. Ge, J. Kurose, Y. Liu, and D. Towsley. Optimal routing with multiple traffic matrices: Tradeoff between average case and worst case performance. In Proc. International Conference on Network Protocols, Nov. 2005. Google ScholarDigital Library
Index Terms
- Rehoming edge links for better traffic engineering
Recommendations
Deriving traffic demands for operational IP networks: methodology and experience
Engineering a large IP backbone network without an accurate network-wide view of the traffic demands is challenging. Shifts in user behavior, changes in routing policies, and failures of network elements can result in significant (and sudden) ...
Performance analysis of a traffic engineering solution for multilayer networks based on the GMPLS paradigm
This paper presents an integrated traffic engineering (TE) system for new generation multilayer networks based on the generalized multiprotocol label switching (GMPLS) paradigm, and reports the performance analysis of such a system. The proposed TE ...
Internet traffic engineering using multi-protocol label switching (MPLS)
Special issue: Towards a new internet architectureWith the rising popularity of the Internet there have arisen corresponding requirements for network reliability, efficiency, and service quality. Internet service providers are responding to these developments by critically examining every aspect of ...
Comments