Skip to main content

A First Measurement with BGP Egress Peer Engineering

Part of the Lecture Notes in Computer Science book series (LNCS,volume 13210)


This paper reports on measuring the effect of engineering egress traffic to peering ASes using Segment Routing, called BGP-EPE. BGP-EPE can send packets destined to arbitrary prefixes to arbitrary eBGP peers regardless of the BGP path selection. This ability enables us to measure external connectivity from a single AS in various perspectives; for example, does the use of paths other than the BGP best paths improve performance? We conducted an experiment to measure latency to the Internet from an event network, Interop Tokyo ShowNet, where SR-MPLS and BGP-EPE were deployed. Our findings from the experiment show BGP-EPE improves latency for 77% of target prefixes, and peering provides shorter latency than transit. We further show factors on which the degree of improvement depends, e.g., the performance-obliviousness of BGP and the presence of remote peering. Also, we find 91% of peer ASes forwarded packets towards prefixes that the peers did not advertise.


  • BGP egress peer engineering
  • Segment routing
  • Internet latency

This is a preview of subscription content, access via your institution.

Buying options

USD   29.95
Price excludes VAT (USA)
  • DOI: 10.1007/978-3-030-98785-5_9
  • Chapter length: 17 pages
  • Instant PDF download
  • Readable on all devices
  • Own it forever
  • Exclusive offer for individuals only
  • Tax calculation will be finalised during checkout
USD   84.99
Price excludes VAT (USA)
  • ISBN: 978-3-030-98785-5
  • Instant PDF download
  • Readable on all devices
  • Own it forever
  • Exclusive offer for individuals only
  • Tax calculation will be finalised during checkout
Softcover Book
USD   109.99
Price excludes VAT (USA)
Fig. 1.
Fig. 2.
Fig. 3.
Fig. 4.
Fig. 5.
Fig. 6.
Fig. 7.
Fig. 8.
Fig. 9.
Fig. 10.
Fig. 11.
Fig. 12.
Fig. 13.
Fig. 14.
Fig. 15.
Fig. 16.
Fig. 17.


  1. 1.

    We used an unanonymized version of the trace with a responsible person’s consent.

  2. 2.

    MAWI Traffic Archive FAQ,


  1. Alexa top 1 million sites.

  2. Interop Tokyo 2021 (2021).

  3. Interop Tokyo 2021 ShowNet (2021).

  4. Ahmed, A., Shafiq, Z., Bedi, H., Khakpour, A.: Peering vs. transit: performance comparison of peering and transit interconnections. In: 2017 IEEE 25th International Conference on Network Protocols (ICNP), pp. 1–10 (2017).

  5. Akella, A., Maggs, B., Seshan, S., Shaikh, A.: On the performance benefits of multihoming route control. IEEE/ACM Trans. Netw. 16(1), 91–104 (2008)

    CrossRef  Google Scholar 

  6. Apostolaki, M., Singla, A., Vanbever, L.: Performance-driven internet path selection. In: Proceedings of the Symposium on SDN Research, SOSR ’21. Association for Computing Machinery, New York (2021).

  7. Arnold, T., et al.: Beating BGP is harder than we thought. In: Proceedings of the 18th ACM Workshop on Hot Topics in Networks, pp. 9b–16. HotNets ’19. Association for Computing Machinery, New York (2019).

  8. Augustin, B., et al.: Avoiding traceroute anomalies with Paris traceroute. In: Proceedings of the 6th ACM SIGCOMM Conference on Internet Measurement, IMC ’06, pp. 153–158. Association for Computing Machinery, New York (2006).

  9. CAIDA: As rank: A ranking of the largest autonomous systems (as) in the internet.

  10. Castro, I., Cardona, J.C., Gorinsky, S., Francois, P.: Remote peering: more peering without internet flattening. In: Proceedings of the 10th ACM International on Conference on Emerging Networking Experiments and Technologies, CoNEXT ’14, pp. 185–198. Association for Computing Machinery, New York (2014).

  11. Chen, E., Sangli, R.S.: Avoid BGP best path transitions from one external to another. RFC 5004 (2007).

  12. Cho, K., Mitsuya, K., Kato, A.: Traffic data repository at the wide project. In: Proceedings of the Annual Conference on USENIX Annual Technical Conference, ATEC ’00, p. 51. USENIX Association, USA (2000)

    Google Scholar 

  13. Claypool, M., Claypool, K.: Latency can kill: precision and deadline in online games, pp. 215–222. Association for Computing Machinery, New York (2010).

  14. Dac Duy Nguyen, H., Secci, S.: LISP-EC: enhancing lisp with egress control. In: 2016 IEEE Conference on Standards for Communications and Networking (CSCN), pp. 1–7 (2016).

  15. Faratin, P., Clark, D., Bauer, S., Lehr, W., Gilmore, P., Berger, A.: The growing complexity of internet interconnection. Commun. Strat. 1, 51–72 (2008)

    Google Scholar 

  16. Feamster, N., Borkenhagen, J., Rexford, J.: Guidelines for interdomain traffic engineering. SIGCOMM Comput. Commun. Rev. 33(5), 19–30 (2003).

    CrossRef  Google Scholar 

  17. Filsfils, C., Previdi, S., Dawra, G., Aries, E., Afanasiev, D.: Segment routing centralized BGP egress peer engineering. RFC 9087 (2021).

  18. Filsfils, C., Previdi, S., Ginsberg, L., Decraene, B., Litkowski, S., Shakir, R.: Segment routing architecture. RFC 8402 (2018).

  19. Goldenberg, D.K., Qiuy, L., Xie, H., Yang, Y.R., Zhang, Y.: Optimizing cost and performance for multihoming. SIGCOMM Comput. Commun. Rev. 34(4), 79–92 (2004)

    CrossRef  Google Scholar 

  20. Huston, G.: AS Names.

  21. Jager, M.: Securing ixp connectivity. APINIC 34 (2012).

  22. Juniper Networks: Containerized routing protocol daemon (CRPD) (2021).

  23. Khan, F.: The cost of latency—digital realty (2015).

  24. Luckie, M.: Scamper: a scalable and extensible packet prober for active measurement of the internet. In: Proceedings of the 10th ACM SIGCOMM Conference on Internet Measurement, IMC ’10, pp. 239–245. Association for Computing Machinery, New York (2010).

  25. Norton, W.: DrPeering white paper - the art of peering: the peering playbook, 7. partial transit (regional) (2010).

  26. PCI-SIG: Single root i/o virtualization and sharing specification revision 1.1 (2010).

  27. Rekhter, Y., Hares, S., Li, T.: A Border Gateway Protocol 4 (BGP-4). RFC 4271 (2006).

  28. Schlinker, B., et al.: Engineering egress with edge fabric: steering oceans of content to the world. In: Proceedings of the Conference of the ACM Special Interest Group on Data Communication, SIGCOMM ’17, pp. 418–431. Association for Computing Machinery, New York (2017).

  29. StefPrevidi, S., Talaulikar, K., Filsfils, Clarence Filand Patel, K., Ray, S., Dong, J.: Border Gateway Protocol - Link State (BGP-LS) Extensions for Segment Routing BGP Egress Peer Engineering. RFC 9086 (2021).

  30. The ANT Lab: IP Address Space Hitlists.

  31. Tsurumaki, S.: How do we improve internet connectivity outside major cities? A Japanese approach. APNIC Blog (2021).

  32. Ventre, P.L., et al.: Segment routing: a comprehensive survey of research activities, standardization efforts, and implementation results. IEEE Commun. Surv. Tutor. 23(1), 182–221 (2021).

    CrossRef  Google Scholar 

  33. Yap, K.K., et al.: Taking the edge off with espresso: scale, reliability and programmability for global internet peering. In: Proceedings of the Conference of the ACM Special Interest Group on Data Communication, SIGCOMM ’17, pp. 432–445. Association for Computing Machinery, New York (2017).

Download references


We would like to thank our shepherd Dr. Marcel Flores and the anonymous reviewers for their insightful feedback. We also thank all the people involved in Interop Tokyo ShowNet 2021.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Ryo Nakamura .

Editor information

Editors and Affiliations



A Other Metrics for Representative RTTs

We chose the minimum, not median, RTTs for each prefix to avoid statistical errors. Figure 18a shows the median version of Fig. 5. As shown, there is no significant difference between minimum and median in a broad view. However, Fig. 19, which shows the median version of improved latency with peering (Fig. 11c, 12c, 13b, and 14b), includes such errors. Accidentally overestimated RTTs cause inaccurate latency differences between peering and transit.

Fig. 18.
figure 18

CDF of RTTs to target prefixes or addresses.

Fig. 19.
figure 19

Peering vs. transit with median RTTs on a per-prefix basis.

Throughout the paper, we summarized RTTs by prefixes in the BGP table. Figure 18b shows the not-summarized version, which means per-address RTT, of Fig. 5. Differences between the figures, e.g., the per-address version has a higher rate with RTTs under 100 ms, arises from a bias on the distribution of target addresses. In addition, Fig. 20 shows the per-address version of peering versus transit (Fig. 11c, 12c, 13b, and 14b). The figure shows results similar to the per-prefix versions. This is because the number of prefixes advertised from the peers was small (3526 unique prefixes). As a result, there were no especially large prefixes that accommodated many target addresses.

Fig. 20.
figure 20

Peering vs. transit with minimum RTTs on per-address basis.

Rights and permissions

Reprints and Permissions

Copyright information

© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG

About this paper

Verify currency and authenticity via CrossMark

Cite this paper

Nakamura, R., Shimizu, K., Kamata, T., Pelsser, C. (2022). A First Measurement with BGP Egress Peer Engineering. In: Hohlfeld, O., Moura, G., Pelsser, C. (eds) Passive and Active Measurement. PAM 2022. Lecture Notes in Computer Science, vol 13210. Springer, Cham.

Download citation

  • DOI:

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-030-98784-8

  • Online ISBN: 978-3-030-98785-5

  • eBook Packages: Computer ScienceComputer Science (R0)