Heterogeneous Gossip

  • Davide Frey
  • Rachid Guerraoui
  • Anne-Marie Kermarrec
  • Boris Koldehofe
  • Martin Mogensen
  • Maxime Monod
  • Vivien Quéma
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5896)

Abstract

Gossip-based information dissemination protocols are considered easy to deploy, scalable and resilient to network dynamics. Load-balancing is inherent in these protocols as the dissemination work is evenly spread among all nodes. Yet, large-scale distributed systems are usually heterogeneous with respect to network capabilities such as bandwidth. In practice, a blind load-balancing strategy might significantly hamper the performance of the gossip dissemination.

This paper presents HEAP, HEterogeneity-Aware gossip Protocol, where nodes dynamically adapt their contribution to the gossip dissemination according to their bandwidth capabilities. Using a continuous, itself gossip-based, approximation of relative bandwidth capabilities, HEAP dynamically leverages the most capable nodes by increasing their fanout, while decreasing by the same proportion that of less capable nodes. HEAP preserves the simple and proactive (churn adaptation) nature of gossip, while significantly improving its effectiveness. We extensively evaluate HEAP in the context of a video streaming application on a testbed of 270 PlanetLab nodes. Our results show that HEAP significantly improves the quality of the streaming over standard homogeneous gossip protocols, especially when the stream rate is close to the average available bandwidth.

References

  1. 1.
    Birman, K., Hayden, M., Ozkasap, O., Xiao, Z., Budiu, M., Minsky, Y.: Bimodal Multicast. TOCS 17(2), 41–88 (1999)CrossRefGoogle Scholar
  2. 2.
    Bishop, M., Rao, S., Sripanidulchai, K.: Considering Priority in Overlay Multicast Protocols under Heterogeneous Environments. In: Proc. of INFOCOM (2006)Google Scholar
  3. 3.
    Bonald, T., Massoulié, L., Mathieu, F., Perino, D., Twigg, A.: Epidemic Live Streaming: Optimal Performance Trade-Offs. In: Proc. of SIGMETRICS (2008)Google Scholar
  4. 4.
    Castro, M., Druschel, P., Kermarrec, A.-M., Nandi, A., Rowstron, A., Singh, A.: SplitStream: High-Bandwidth Multicast in Cooperative Environments. In: Proc. of SOSP (2003)Google Scholar
  5. 5.
    Chu, Y.-H., Rao, S., Zhang, H.: A Case for End System Multicast. JSAC 20(8), 1456–1471 (2000)Google Scholar
  6. 6.
    Demers, A., Greene, D., Hauser, C., Irish, W., Larson, J., Shenker, S., Sturgis, H., Swinehart, D., Terry, D.: Epidemic Algorithms for Replicated Database Maintenance. In: Proc. of PODC (1987)Google Scholar
  7. 7.
    Deshpande, M., Xing, B., Lazardis, I., Hore, B., Venkatasubramanian, N., Mehrotra, S.: CREW: A Gossip-based Flash-Dissemination System. In: Proc. of ICDCS (2006)Google Scholar
  8. 8.
    Eugster, P., Guerraoui, R., Handurukande, S., Kermarrec, A.-M., Kouznetsov, P.: Lightweight Probabilistic Broadcast. TOCS 21(4), 341–374 (2003)CrossRefGoogle Scholar
  9. 9.
    Frey, D., Guerraoui, R., Kermarrec, A.-M., Monod, M., Quéma, V.: Stretching Gossip with Live Streaming. In: Proc. of DSN (2009)Google Scholar
  10. 10.
    Garbinato, B., Pedone, F., Schmidt, R.: An Adaptive Algorithm for Efficient Message Diffusion in Unreliable Environments. In: Proc. of DSN (2004)Google Scholar
  11. 11.
    Guerraoui, R., Huguenin, K., Kermarrec, A.-M., Monod, M.: On Tracking Freeriders in Gossip Protocols. In: Proc. of P2P (2009)Google Scholar
  12. 12.
    Hei, X., Liang, C., Liang, J., Liu, Y., Ross, K.: A Measurement Study of a Large-Scale P2P IPTV System. TMM 9(8), 1672–1687 (2007)Google Scholar
  13. 13.
    Jelasity, M., Montresor, A., Babaoglu, O.: Gossip-Based Aggregation in Large Dynamic Networks. TOCS 23(3), 219–252 (2005)CrossRefGoogle Scholar
  14. 14.
    Jenkins, K., Hopkinson, K., Birman, K.: A Gossip Protocol for Subgroup Multicast. In: Proc. of ICDCS Workshops (2001)Google Scholar
  15. 15.
    Kermarrec, A.-M., Massoulié, L., Ganesh, A.: Probabilistic Reliable Dissemination in Large-Scale Systems. TPDS 14(3), 248–258 (2003)Google Scholar
  16. 16.
    Kyasanur, P., Choudhury, R.R., Gupta, I.: Smart Gossip: An Adaptive Gossip-based Broadcasting Service for Sensor Networks. In: Proc. of MASS (2006)Google Scholar
  17. 17.
    Li, B., Qu, Y., Keung, Y., Xie, S., Lin, C., Liu, J., Zhang, X.: Inside the New Coolstreaming: Principles, Measurements and Performance Implications. In: Proc. of INFOCOM (2008)Google Scholar
  18. 18.
    Li, H., Clement, A., Marchetti, M., Kapritsos, M., Robinson, L., Alvisi, L., Dahlin, M.: FlightPath: Obedience vs Choice in Cooperative Services. In: Proc. of OSDI (2008)Google Scholar
  19. 19.
    Li, H., Clement, A., Wong, E., Napper, J., Roy, I., Alvisi, L., Dahlin, M.: BAR Gossip. In: Proc. of OSDI (2006)Google Scholar
  20. 20.
    Liang, C., Guo, Y., Liu, Y.: Is Random Scheduling Sufficient in P2P Video Streaming? In: Proc. of ICDCS (2008)Google Scholar
  21. 21.
    Ma, L., Ooi, W.: Congestion Control in Distributed Media Streaming. In: Proc. of INFOCOM (2007)Google Scholar
  22. 22.
    Magharei, N., Rejaie, R.: PRIME: Peer-to-Peer Receiver-drIven MEsh-based Streaming. In: Proc. of INFOCOM (2007)Google Scholar
  23. 23.
    Pai, V., Kumar, K., Tamilmani, K., Sambamurthy, V., Mohr, A.: Chainsaw: Eliminating Trees from Overlay Multicast. In: Castro, M., van Renesse, R. (eds.) IPTPS 2005. LNCS, vol. 3640, pp. 127–140. Springer, Heidelberg (2005)CrossRefGoogle Scholar
  24. 24.
    Picconi, F., Massoulié, L.: Is There a Future for Mesh-Based live Video Streaming? In: Proc. of P2P (2008)Google Scholar
  25. 25.
  26. 26.
    Spring, N., Peterson, L., Bavier, A., Pai, V.: Using Planetlab for Network Research: Myths, Realities, and Best Practices. OSR 40(1), 17–24 (2006)Google Scholar
  27. 27.
    Sung, Y.-W., Bishop, M., Rao, S.: Enabling Contribution Awareness in an Overlay Broadcasting System. CCR 36(4), 411–422 (2006)Google Scholar
  28. 28.
    van Renesse, R., Birman, K., Vogels, W.: Astrolabe: A Robust and Scalable Technology for Distributed System Monitoring, Management, and Data Mining. TOCS 21(2), 164–206 (2003)CrossRefGoogle Scholar
  29. 29.
    Venkataraman, V., Yoshida, K., Francis, P.: Chunkyspread: Heterogeneous Unstructured Tree-Based Peer to Peer Multicast. In: Proc. of ICNP (2006)Google Scholar
  30. 30.
    Vishnumurthy, V., Francis, P.: On Heterogeneous Overlay Construction and Random Node Selection in Unstructured P2P Networks. In: Proc. of INFOCOM (2006)Google Scholar
  31. 31.
    Wang, W., Jin, C., Jamin, S.: Network Overlay Construction under Limited End-to-End Reachability. In: Proc. of INFOCOM (2005)Google Scholar
  32. 32.
    Widmer, J., Handley, M.: Extending Equation-based Congestion Control to Multicast Applications. In: Proc. of SIGCOMM (2001)Google Scholar
  33. 33.
  34. 34.
    Zhang, B., Wang, W., Jamin, S., Massey, D., Zhang, L.: Universal IP multicast delivery. Computer Networks 50(6), 781–806 (2006)CrossRefGoogle Scholar
  35. 35.
    Zhang, M., Zhang, Q., Sun, L., Yang, S.: Understanding the Power of Pull-Based Streaming Protocol: Can We Do Better? JSAC 25(9), 1678–1694 (2007)Google Scholar

Copyright information

© IFIP International Federation for Information Processing 2009

Authors and Affiliations

  • Davide Frey
    • 2
  • Rachid Guerraoui
    • 1
  • Anne-Marie Kermarrec
    • 2
  • Boris Koldehofe
    • 3
  • Martin Mogensen
    • 4
  • Maxime Monod
    • 1
  • Vivien Quéma
    • 5
  1. 1.Ecole Polytechnique Fédérale de LausanneSwitzerland
  2. 2.INRIA Rennes-Bretagne AtlantiqueFrance
  3. 3.University of StuttgartGermany
  4. 4.University of AarhusDenmark
  5. 5.CNRSFrance

Personalised recommendations