Robust and efficient membership management in large-scale dynamic networks

Epidemic protocols are a bio-inspired communication and computation paradigm for large-scale networked systems based on randomised communication. These protocols rely on a membership service to build decentralised and random overlay topologies. In large-scale, dynamic network environments, node churn and failures may have a detrimental effect on the structure of the overlay topologies with negative impact on the efficiency and the accuracy of applications. Most importantly, there exists the risk of a permanent loss of global connectivity that would prevent the correct convergence of applications. This work investigates to what extent a dynamic network environment may negatively affect the performance of Epidemic membership protocols. A novel Enhanced Expander Membership Protocol (EMP+) based on the expansion properties of graphs is presented. The proposed protocol is evaluated against other membership protocols and the comparative analysis shows that EMP+ can support faster application convergence and is the first membership protocol to provide robustness against global network connectivity problems.

[1]  Giuseppe Di Fatta,et al.  Expander Graph Quality Optimisation in Randomised Communication , 2014, 2014 IEEE International Conference on Data Mining Workshop.

[2]  Giancarlo Fortino,et al.  Epidemic K-Means Clustering , 2011, 2011 IEEE 11th International Conference on Data Mining Workshops.

[3]  George Pavlou,et al.  An adaptive clustering approach for the management of dynamic systems , 2005, IEEE Journal on Selected Areas in Communications.

[4]  Giancarlo Fortino,et al.  Fault tolerant decentralised K-Means clustering for asynchronous large-scale networks , 2013, J. Parallel Distributed Comput..

[5]  Jean Mayo,et al.  MuON: Epidemic based mutual anonymity in unstructured P2P networks , 2008, Comput. Networks.

[6]  Hoda Mashayekhi,et al.  GDCluster: A General Decentralized Clustering Algorithm , 2015, IEEE Transactions on Knowledge and Data Engineering.

[7]  Giancarlo Fortino,et al.  Gossiping-Based AODV for Wireless Sensor Networks , 2013, 2013 IEEE International Conference on Systems, Man, and Cybernetics.

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

[9]  Anne-Marie Kermarrec,et al.  Gossip-based peer sampling , 2007, TOCS.

[10]  G. Fortino,et al.  Symmetric Push-Sum Protocol for Decentralised Aggregation , 2011, AP2PS 2011.

[11]  Giuseppe Di Fatta,et al.  Connectivity Recovery in Epidemic Membership Protocols , 2015, IDCS.

[12]  Sanjay Ranka,et al.  Aggregation methods for large-scale sensor networks , 2008, TOSN.

[13]  Soummya Kar,et al.  Gossip Algorithms for Distributed Signal Processing , 2010, Proceedings of the IEEE.

[14]  Valentin Cristea,et al.  Epidemic-Style Global Load Monitoring in Large-Scale Overlay Networks , 2010, 2010 International Conference on P2P, Parallel, Grid, Cloud and Internet Computing.

[15]  Abbas Jamalipour,et al.  An Epidemic P2P Content Search Mechanism for Intermittently Connected Mobile Ad Hoc Networks , 2009, GLOBECOM 2009 - 2009 IEEE Global Telecommunications Conference.

[16]  Stephen A. Jarvis,et al.  Peer sampling with improved accuracy , 2009, Peer Peer Netw. Appl..

[17]  Giuseppe Di Fatta,et al.  Expansion Quality of Epidemic Protocols , 2014, IDC.

[18]  Piet Van Mieghem,et al.  Modeling gossip-based content dissemination and search in distributed networking , 2011, Comput. Commun..

[19]  N. Linial,et al.  Expander Graphs and their Applications , 2006 .

[20]  Giuseppe Di Fatta,et al.  Scalable and Fault Tolerant Failure Detection and Consensus , 2015, EuroMPI.

[21]  Robbert van Renesse,et al.  A Gossip-Style Failure Detection Service , 2009 .

[22]  Jean-Marie Gorce,et al.  Joint Estimation and Gossip Averaging for Sensor Network Applications , 2010, IEEE Transactions on Automatic Control.

[23]  Márk Jelasity,et al.  PeerSim: A scalable P2P simulator , 2009, 2009 IEEE Ninth International Conference on Peer-to-Peer Computing.

[24]  Michael Lang,et al.  A gossip-based approach to exascale system services , 2013, ROSS '13.

[25]  Wilfried N. Gansterer,et al.  Fault Tolerance Properties of Gossip-Based Distributed Orthogonal Iteration Methods , 2013, ICCS.