Bandwidth Prediction in the Face of Asymmetry

  • Sven Schober
  • Stefan Brenner
  • Rüdiger Kapitza
  • Franz J. Hauck
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7891)


An increasing number of networked applications, like video conference and video-on-demand, benefit from knowledge about Internet path measures like available bandwidth. Server selection and placement of infrastructure nodes based on accurate information about network conditions help to improve the quality-of-service of these systems. Acquiring this knowledge usually requires fully-meshed ad-hoc measurements. These, however, introduce a large overhead and a possible delay in communication establishment. Thus, prediction-based approaches like Sequoia have been proposed, which treat path properties as a semimetric and embed them onto trees, leveraging labelling schemes to predict distances between hosts not measured before. In this paper, we identify asymmetry as a cause of serious distortion in these systems causing inaccurate prediction. We study the impact of asymmetric network conditions on the accuracy of existing tree-embedding approaches, and present direction-aware embedding, a novel scheme that separates upstream from downstream properties of hosts and significantly improves the prediction accuracy for highly asymmetric datasets. This is achieved by embedding nodes for each direction separately and constraining the distance calculation to inversely labelled nodes. We evaluate the effectiveness and trade-offs of our approach using synthetic as well as real-world datasets.


Asymmetric bandwidth prediction tree embedding 


  1. 1.
    Abraham, I., Balakrishnan, M., Kuhn, F., Malkhi, D., Ramasubramanian, V., Talwar, K.: Reconstructing approximate tree metrics. In: Proc. of the 26th Ann. ACM Symp. on Princ. of Distr. Comp., pp. 43–52 (August 2007)Google Scholar
  2. 2.
    Beaumont, O., Eyraud-Dubois, L., Won, Y.J.: Using the last-mile model as a distributed scheme for available bandwidth prediction. In: Proc. of the 17th Int. Conf. on Par. Proc., pp. 103–116 (2011)Google Scholar
  3. 3.
    Chun, B., Culler, D., Roscoe, T., Bavier, A., Peterson, L., Wawrzoniak, M., Bowman, M.: Planetlab: an overlay testbed for broad-coverage services. SIGCOMM Comput. Commun. Rev. 33(3), 3–12 (2003)CrossRefGoogle Scholar
  4. 4.
    Dabek, F., Cox, R., Kaashoek, F., Morris, R.: Vivaldi: a decentralized network coordinate system. SIGCOMM Comput. Commun. Rev. 34(4), 15–26 (2004)CrossRefGoogle Scholar
  5. 5.
    Donnet, B., Gueye, B., Kaafar, M.: A survey on network coordinates systems, design, and security. IEEE Comm. Surveys Tutorials 12(4), 488–503 (2010)CrossRefGoogle Scholar
  6. 6.
    Dovrolis, C., Ramanathan, P., Moore, D.: Packet-dispersion techniques and a capacity-estimation methodology. Trans. on Netw. 12(6), 963–977 (2004)CrossRefGoogle Scholar
  7. 7.
    Elser, B., Förschler, A., Fuhrmann, T.: Tuning vivaldi: Achieving increased accuracy and stability. In: Spyropoulos, T., Hummel, K.A. (eds.) IWSOS 2009. LNCS, vol. 5918, pp. 174–184. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  8. 8.
    Francis, P., Jamin, S., Jin, C., Jin, Y., Raz, D., Shavitt, Y., Zhang, L.: Idmaps: a global internet host distance estimation service. Trans. on Netw. 9(5), 525–540 (2001)CrossRefGoogle Scholar
  9. 9.
    Haddow, T., Ho, S.W., Ledlie, J., Lumezanu, C., Draief, M., Pietzuch, P.: On the feasibility of bandwidth detouring. In: Spring, N., Riley, G.F. (eds.) PAM 2011. LNCS, vol. 6579, pp. 81–91. Springer, Heidelberg (2011)CrossRefGoogle Scholar
  10. 10.
    Hu, N., Steenkiste, P.: Exploiting internet route sharing for large scale available bandwidth estimation. In: Proc. of the 5th ACM SIGCOMM Conf. on Internet Meas., p. 16 (2005)Google Scholar
  11. 11.
    Lakshminarayanan, K., Padmanabhan, V.N.: Some findings on the network performance of broadband hosts. In: Proc. of the 3rd ACM SIGCOMM Conf. on Internet Meas., pp. 45–50 (2003)Google Scholar
  12. 12.
    Liao, Y., Du, W., Geurts, P., Leduc, G.: Dmfsgd: A decentralized matrix factorization algorithm for network distance prediction. Trans. on Netw. PP(99), 1 (2012)CrossRefGoogle Scholar
  13. 13.
    Liao, Y., Geurts, P., Leduc, G.: Network distance prediction based on decentralized matrix factorization. In: Proc. of the 9th IFIP TC 6 Int. Conf. on Netw., pp. 15–26 (2010)Google Scholar
  14. 14.
    Liu, S., Zhang-Shen, R., Jiang, W., Rexford, J., Chiang, M.: Performance bounds for peer-assisted live streaming. SIGMETRICS Perform. Eval. Rev. 36(1), 313–324 (2008)CrossRefGoogle Scholar
  15. 15.
    Lumezanu, C., Baden, R., Levin, D., Spring, N., Bhattacharjee, B.: Symbiotic relationships in internet routing overlays. In: Proc. of the 6th USENIX Symp. on Netw. Sys. Des. and Impl., NSDI 2009, pp. 467–480 (2009)Google Scholar
  16. 16.
    Madhyastha, H.V., Isdal, T., Piatek, M., Dixon, C., Anderson, T., Krishnamurthy, A., Venkataramani, A.: iplane: an information plane for distributed services. In: Proc. of the 7th OSDI Conf., pp. 367–380 (2006)Google Scholar
  17. 17.
    Mao, Y., Saul, L., Smith, J.: Ides: An internet distance estimation service for large networks. J. on Sel. Areas in Comm. 24(12), 2273–2284 (2006)CrossRefGoogle Scholar
  18. 18.
    Ng, T.S.E., Zhang, H.: Predicting internet network distance with coordinates-based approaches. In: Proc. IEEE INFOCOM, vol. 1, pp. 170–179 (2002)Google Scholar
  19. 19.
    Paxson, V.: End-to-end routing behavior in the internet. SIGCOMM Comput. Commun. Rev. 36(5), 41–56 (2006)CrossRefGoogle Scholar
  20. 20.
    Peleg, D.: Proximity-preserving labeling schemes. J. Graph Theory 33(3), 167–176 (2000)MathSciNetzbMATHCrossRefGoogle Scholar
  21. 21.
    Ramasubramanian, V., Malkhi, D., Kuhn, F., Balakrishnan, M., Gupta, A., Akella, A.: On the treeness of internet latency and bandwidth. In: Proc. of the 11th Int. Joint Conf. on Meas. and Modeling of Comp. Sys., pp. 61–72 (2009)Google Scholar
  22. 22.
    Ribeiro, V., Riedi, R., Baraniuk, R., Navratil, J., Cottrell, L.: pathchirp: Efficient available bandwidth estimation for network paths. In: Passive and Active Meas. Workshop (2003)Google Scholar
  23. 23.
    Shavitt, Y., Tankel, T.: Hyperbolic embedding of internet graph for distance estimation and overlay construction. Trans. on Netw. 16(1), 25–36 (2008)CrossRefGoogle Scholar
  24. 24.
    Song, S.: Decentralized pairwise bandwidth prediction, Unknown Date,
  25. 25.
    Song, S., Keleher, P., Bhattacharjee, B., Sussman, A.: Decentralized, accurate, and low-cost network bandwidth prediction. In: Proc. IEEE INFOCOM, pp. 6–10 (April 2011)Google Scholar
  26. 26.
    Strauss, J., Katabi, D., Kaashoek, F.: A measurement study of available bandwidth estimation tools. In: Proc. of the 3rd ACM SIGCOMM Conf. on Internet Meas., pp. 39–44 (2003)Google Scholar
  27. 27.
    Xing, C., Chen, M., Yang, L.: Predicting available bandwidth of internet path with ultra metric space-based approaches. In: Proc. of IEEE GLOBECOM, pp. 1–6 (2009)Google Scholar

Copyright information

© IFIP International Federation for Information Processing 2013

Authors and Affiliations

  • Sven Schober
    • 1
  • Stefan Brenner
    • 2
  • Rüdiger Kapitza
    • 2
  • Franz J. Hauck
    • 1
  1. 1.Institute of Distributed SystemsUniversity of UlmGermany
  2. 2.TU BraunschweigGermany

Personalised recommendations