Mean-Field Analysis for the Evaluation of Gossip Protocols

Gossip protocols are designed to operate in very large, decentralised networks. A node in such a network bases its decision to interact (gossip) with another node on its partial view of the global system. Because of the size of these networks, analysis of gossip protocols is mostly done using simulations, that tend to be expensive in computation time and memory consumption. We employ mean-¿eld approximation for an analytical evaluation of gossip protocols. Nodes in the network are represented by small identical stochastic models. Joining all nodes would result in an enormous stochastic process. If the number of nodes goes to in¿nity, however, mean-¿eld analysis allows us to replace this intractably large stochastic process by a small deterministic process. This process approximates the behaviour of very large gossip networks, and can be evaluated using simple matrix-vector multiplications.

[1]  Anne-Marie Kermarrec,et al.  Gossiping in distributed systems , 2007, OPSR.

[2]  François Bonnet,et al.  Brief Announcement: Performance Analysis of Cyclon, an Inexpensive Membership Management for Unstructured P2P Overlays , 2006, DISC.

[3]  John E. Hopcroft,et al.  Correctness of a gossip based membership protocol , 2005, PODC '05.

[4]  Konrad Iwanicki,et al.  Gossip-Based Clock Synchronization for Large Decentralized Systems , 2006, SelfMan.

[5]  Dharma P. Agrawal,et al.  Dynamic probabilistic broadcasting in MANETs , 2005, J. Parallel Distributed Comput..

[6]  Maarten van Steen,et al.  CYCLON: Inexpensive Membership Management for Unstructured P2P Overlays , 2005, Journal of Network and Systems Management.

[7]  M. V. Steen,et al.  Newscast Computing , 2003 .

[8]  Alcherio Martinoli,et al.  Modeling Swarm Robotic Systems: a Case Study in Collaborative Distributed Manipulation , 2004, Int. J. Robotics Res..

[9]  Gian Paolo Jesi,et al.  Exploring the interdisciplinary connections of gossip-based systems , 2007, OPSR.

[10]  N. H. Lee,et al.  Fluid and Brownian approximations for an Internet congestion control model , 2004, 2004 43rd IEEE Conference on Decision and Control (CDC) (IEEE Cat. No.04CH37601).

[11]  Miklós Telek,et al.  Analysis of Large Scale Interacting Systems by Mean Field Method , 2008, 2008 Fifth International Conference on Quantitative Evaluation of Systems.

[12]  F. Baccelli,et al.  Metastable Regimes for multiplexed TCP flows , 2004 .

[13]  Márk Jelasity,et al.  Gossip-based aggregation in large dynamic networks , 2005, TOCS.

[14]  Doug Terry,et al.  Epidemic algorithms for replicated database maintenance , 1988, OPSR.

[15]  Armand M. Makowski,et al.  Limit behavior of ECN/RED gateways under a large number of TCP flows , 2003, IEEE INFOCOM 2003. Twenty-second Annual Joint Conference of the IEEE Computer and Communications Societies (IEEE Cat. No.03CH37428).

[16]  David Starobinski,et al.  Data Dissemination in Wireless Broadcast Channels: Network Coding or Cooperation , 2007, 2007 41st Annual Conference on Information Sciences and Systems.

[17]  N. Vvedenskaya,et al.  Dobrushin's Mean-Field Approximation for a Queue With Dynamic Routing , 1997 .

[18]  Boudewijn R. Haverkort,et al.  Mean-Field Analysis for the Evaluation of Gossip Protocols , 2009, QEST.

[19]  Marta Z. Kwiatkowska,et al.  Analysis of a gossip protocol in PRISM , 2008, PERV.

[20]  Boudewijn R. Haverkort,et al.  Formal analysis techniques for gossiping protocols , 2007, OPSR.

[21]  Scott Shenker,et al.  Epidemic algorithms for replicated database maintenance , 1988, OPSR.

[22]  Laurent Massoulié,et al.  Integrating streaming and file-transfer Internet traffic: fluid and diffusion approximations , 2007, Queueing Syst. Theory Appl..

[23]  François Baccelli,et al.  A mean-field analysis of short lived interacting TCP flows , 2004, SIGMETRICS '04/Performance '04.

[24]  François Baccelli,et al.  A mean-field model for multiple TCP connections through a buffer implementing RED , 2002, Perform. Evaluation.

[25]  Martin L. Puterman,et al.  Markov Decision Processes: Discrete Stochastic Dynamic Programming , 1994 .

[26]  Y. Suhov,et al.  Dobrushin's approach to queueing network theory , 1996 .

[27]  Yiqiang Q. Zhao,et al.  Balancing Queues by Mean Field Interaction , 2005, Queueing Syst. Theory Appl..

[28]  Anne-Marie Kermarrec,et al.  The Peer Sampling Service: Experimental Evaluation of Unstructured Gossip-Based Implementations , 2004, Middleware.

[29]  François Baccelli,et al.  HTTP turbulence , 2006, Networks Heterog. Media.

[30]  G. Fayolle,et al.  Models for transporation networks , 1997 .

[31]  Jean-Yves Le Boudec,et al.  A Generic Mean Field Convergence Result for Systems of Interacting Objects , 2007, Fourth International Conference on the Quantitative Evaluation of Systems (QEST 2007).

[32]  M. van Steen,et al.  A Gossip-based Distributed News Service for Wireless Mesh Networks , 2006 .

[33]  U. Rieder,et al.  Markov Decision Processes , 2010 .

[34]  Janina Mincer-Daszkiewicz.  Gossip-Based Dissemination of Time , 2005 .