Abstract

Border Gateway Protocol (BGP) convergence time is a primary determinant of service availability in service provider (SP) networks following link or routing-session failures. Slow convergence introduces transient traffic blackholes that degrade throughput and violate service-level agreements (SLAs). This paper presents a reproducible NS-3 v3.43 simulation study that models BGP-like failover behavior through event-driven routing-table updates parameterized by hold-timer, MRAI, and BFD assumptions informed by RFC timer guidance and explicitly selected propagation and FIB assumptions. All convergence times are explicit design inputs; NS-3 FlowMonitor measures the resulting packet loss and throughput. Four scenarios are evaluated on a dual-homed six-node SP topology: stable routing, default-timer failover, fast-timer failover with BFD, and BFD-triggered fast failover. The model maps assumed blackhole-window durations to packet loss and MTBF/MTTR availability under controlled NS-3 forwarding conditions. Default-timer failover (Tconv = 182.3 s, midpoint hold-timer detection with worst-case MRAI delay) yields 30.5% packet loss and 2.74 nines of availability. Tuned timers with BFD achieve 1.04% loss and 4.21 nines; BFD-triggered fast failover achieves 0.047% loss and 5.55 nines. A sensitivity analysis across four illustrative MTBF assumptions shows that S4 improves from 5.55 to 8.05 nines as MTBF increases from 27.8 hours to one year, while S2 reaches only 5.24 nines at the same annual MTBF. Analytical MRAI and BFD interval sweeps further show that MRAI reduction is the primary optimization lever in BFD-enabled deployments, with the optimum MRAI in the 2–5 s range where marginal availability gains diminish relative to the increase in UPDATE message rates.

References

• Y. Rekhter, T. Li, and S. Hares, “A Border Gateway Protocol 4 (BGP-4),” IETF RFC 4271, Jan. 2006. https://doi.org/ 10.17487/RFC4271
• C. Labovitz, A. Ahuja, A. Bose, and F. Jahanian, “Delayed internet routing convergence,” in Proc. ACM SIGCOMM, Stockholm, Aug. 2000, pp. 175–187. https://doi.org/ 10.1145/347059.347428
• T. G. Griffin and B. J. Premore, “An experimental analysis of BGP convergence time,” in Proc. IEEE ICNP, Riverside, CA, Nov. 2001, pp. 53–61. https://doi.org/10.1109/ICNP. 2001.992898
• Z. M. Mao, R. Govindan, G. Varghese, and R. H. Katz, “Route flap damping exacerbates internet routing convergence,” in Proc. ACM SIGCOMM, Pittsburgh, PA, Aug. 2002, pp. 221–233. https://doi.org/10.1145/633025.633047
• P. Francois, C. Filsfils, J. Evans, and O. Bonaventure, “Achieving sub-second IGP convergence in large IP networks,” ACM SIGCOMM Comput. Commun. Rev., vol. 35, no. 3, pp. 35–44, Jul. 2005. https://doi.org/10.1145/ 1070873.1070877
• D. Katz and D. Ward, “Bidirectional Forwarding Detection (BFD),” IETF RFC 5880, Jun. 2010. https://doi.org/ 10.17487/RFC5880
• A. Sahoo, K. Kant, and P. Mohapatra, “Improving BGP convergence delay for large-scale failures,” in Proc. IEEE/IFIP DSN, Philadelphia, PA, Jun. 2006, pp. 323–332. https://doi.org/10.1109/DSN.2006.41
• M. Milani, M. Segata, L. Baldesi, M. Nesler, R. Lo Cigno, and L. Maccari, “Optimizing MRAI on large scale BGP networks: An emulation-based approach,” Comput. Commun., vol. 228, 2024. https://doi.org/10.1016/j. comcom.2024.107940
• D. Pei, B. Zhang, D. Massey, and L. Zhang, “An analysis of convergence delay in path vector routing protocols,” Comput. Netw., vol. 50, no. 3, pp. 398–421, Feb. 2006. https://doi. org/10.1016/j.comnet.2005.04.013
• J. Haas, “A BGP Cease NOTIFICATION Subcode for Bidirectional Forwarding Detection (BFD),” IETF RFC 9384, Mar. 2023. https://doi.org/10.17487/RFC9384
• A. Atlas and A. Zinin, “Basic specification for IP fast reroute: Loop-free alternates,” IETF RFC 5286, Sep. 2008. https: //doi.org/10.17487/RFC5286
• G. F. Riley and T. R. Henderson, “The ns-3 network simulator,” in Modeling and Tools for Network Simulation, K. Wehrle, M. Gunes, and J. Gross, Eds. Berlin: Springer, 2010, pp. 15–34. https: //doi.org/10.1007/978-3-642-12331-3_2.