Gnutella: Topology Dynamics On Phase Space

In this paper, the topology dynamic of Gnutella is studied through phase space. The dynamic changes in peer degree are studied as a time series in two dimensional phase space which is defined as the number of connected leaves and the number of connected ultras. The reported degrees are concentrated in three special Software related regions that we named as Ultra Stable Region, Leaf Stable Region and Transition Belt. A method is proposed to classify degree traces in phase space into different classes. Connection churn then is studied along with the churn in degree. It shows that the topological structure of Gnutella is rather stable in its connection degree but not the topology itself. The connection drop rate is estimated and the live time of connections is inferred afterwards. M/M/m/m loss queue system is introduced to model the degree keeping process in Gnutella. This model revealed that the degree stable is ensured by large new connection efforts. In other words the stable in topological structure of Gnutella is a results of essential unstable in its topology. That opens a challenge to the basic design philosophy of this network.

[1]  Dimitri P. Bertsekas,et al.  Data Networks , 1986 .

[2]  Ian T. Foster,et al.  Mapping the Gnutella Network: Properties of Large-Scale Peer-to-Peer Systems and Implications for System Design , 2002, ArXiv.

[3]  Hans-Peter Kriegel,et al.  LOF: identifying density-based local outliers , 2000, SIGMOD '00.

[4]  Aleksandar Kuzmanovic,et al.  Denial-of-service resilience in peer-to-peer file sharing systems , 2005, SIGMETRICS '05.

[5]  Priya Mahadevan,et al.  Lessons from Three Views of the Internet Topology , 2005, ArXiv.

[6]  Nicolas Christin,et al.  Content availability, pollution and poisoning in file sharing peer-to-peer networks , 2005, EC '05.

[7]  Stefan Savage,et al.  Understanding Availability , 2003, IPTPS.

[8]  Daniel Stutzbach,et al.  Characterizing files in the modern Gnutella network , 2006, Electronic Imaging.

[9]  Daniel Stutzbach,et al.  Capturing Accurate Snapshots of the Gnutella Network , 2005, Proceedings IEEE INFOCOM 2006. 25TH IEEE International Conference on Computer Communications.

[10]  Rakesh Kumar,et al.  Pollution in P2P file sharing systems , 2005, Proceedings IEEE 24th Annual Joint Conference of the IEEE Computer and Communications Societies..

[11]  Michalis Faloutsos,et al.  Is P2P dying or just hiding? [P2P traffic measurement] , 2004, IEEE Global Telecommunications Conference, 2004. GLOBECOM '04..

[12]  Krishna P. Gummadi,et al.  Measuring and analyzing the characteristics of Napster and Gnutella hosts , 2003, Multimedia Systems.

[13]  Daniel Stutzbach,et al.  Evaluating the Accuracy of Captured Snapshots by Peer-to-Peer Crawlers , 2005, PAM.

[14]  Stefan Saroiu,et al.  A Measurement Study of Peer-to-Peer File Sharing Systems , 2001 .

[15]  Mary K. Vernon,et al.  Characterizing the query behavior in peer-to-peer file sharing systems , 2004, IMC '04.

[16]  Michalis Faloutsos,et al.  Transport layer identification of P2P traffic , 2004, IMC '04.

[17]  Daniel Stutzbach,et al.  Characterizing Unstructured Overlay Topologies in Modern P2P File-Sharing Systems , 2005, IEEE/ACM Transactions on Networking.