How (Not) to Shoot in Your Foot with SDN Local Fast Failover

A Load-Connectivity Tradeoff
  • Michael Borokhovich
  • Stefan Schmid
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8304)

Abstract

This paper studies the resilient routing and (in-band) fast failover mechanisms supported in Software-Defined Networks (SDN). We analyze the potential benefits and limitations of such failover mechanisms, and focus on two main metrics: (1) correctness (in terms of connectivity and loop-freeness) and (2) load-balancing. We make the following contributions. First, we show that in the worst-case (i.e., under adversarial link failures), the usefulness of local failover is rather limited: already a small number of failures will violate connectivity properties under any fast failover policy, even though the underlying substrate network remains highly connected. We then present randomized and deterministic algorithms to compute resilient forwarding sets; these algorithms achieve an almost optimal tradeoff. Our worst-case analysis is complemented with a simulation study.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Borokhovich, M., Schmid, S.: How (not) to shoot in your foot with sdn local fast failover: A load-connectivity tradeoff. Technical Report, arXiv:1309.3150 (2013)Google Scholar
  2. 2.
    Ba, J.F., et al.: On the resilience of routing tables. In: Proc. ACM Symposium on Principles of Distributed Computing (PODC), pp. 237–238 (2012)Google Scholar
  3. 3.
    Lakshminarayanan, K., Caesar, M., Rangan, M., Anderson, T., Shenker, S., Stoica, I.: Achieving convergence-free routing using failure-carrying packets. In: Proc. SIGCOMM, pp. 241–252 (2007)Google Scholar
  4. 4.
    Liu, J., Yan, B., Shenker, S., Schapira, M.: Data-driven network connectivity. In: Proc. HotNets, pp. 8:1–8:6 (2011)Google Scholar
  5. 5.
    Lor, S.S., Landa, R., Rio, M.: Packet re-cycling: eliminating packet losses due to network failures. In: Proc. HotNets, pp. 2:1–2:6 (2010)Google Scholar
  6. 6.
  7. 7.
    Pan, P., Swallow, G., Atlas, A.: Fast reroute extensions to RSVP-TE for LSP tunnels. In: RFC 4090 (2005)Google Scholar
  8. 8.
    Saito, H., Yoshida, M.: An optimal recovery LSP assignment scheme for MPLS fast reroute. In: Proc. NETWORKS (2002)Google Scholar
  9. 9.
    Schmid, S., Suomela, J.: Exploiting locality in distributed sdn control. In: Proc. SIGCOMM Workshop on Hot Topics in Software Defined Networking, HotSDN (2013)Google Scholar
  10. 10.
    Suchara, M., Xu, D., Doverspike, R., Johnson, D., Rexford, J.: Network architecture for joint failure recovery and traffic engineering. In: Proc. ACM SIGMETRICS, pp. 97–108 (2011)Google Scholar
  11. 11.
    Vasseur, J.-P., Pickavet, M., Demeester, P.: Network Recovery: Protection and Restoration of Optical, SONET-SDH, IP, and MPLS. Morgan Kaufmann Publishers Inc. (2004)Google Scholar
  12. 12.
    Wang, D., Li, G.: Efficient distributed bandwidth management for MPLS fast reroute. IEEE/ACM Trans. Netw. (2008)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2013

Authors and Affiliations

  • Michael Borokhovich
    • 1
  • Stefan Schmid
    • 2
  1. 1.Ben-Gurion University of the NegevIsrael
  2. 2.TU Berlin & T-LabsGermany

Personalised recommendations