Measurements of Peer-to-Peer SystemsIntroduction to Peer-to-Peer (P2P) systemsClassification of P2P systemsPopularity of unstructured decentralized P2P networksOutlineGnutella protocol overviewCharacterization of Users of P2P systemsMeasurement MethodologyHost Lifetime analysisLatency analysis (Gnutella)Bottleneck Bandwidth Analysis (Gnutella)Downloads, Uploads and Shared FilesShared files v/s Shared Data (Napster and Gnutella)Degree of Cooperation (Napster)Effect of P2P traffic on underlying networkSlide 16Datasets used for analysisHost distribution analysisHost connectivity analysis (FastTrack)Traffic volume analysisMean bandwidth usage (FastTrack and Direct Connect)Traffic patterns over time (FastTrack)Connection duration and On-time (FastTrack)Peer-to-Peer TopologiesGnutella Network GrowthDistribution of node-to-node shortest pathsAverag node connectivityNode connectivity distributionSearching on the P2P networkTop 20 most popular query typesQuery popularity distributionDeciphering proprietary P2P systemsCharacteristics of Collected TracesFile download distribution by bytesFile size distributionQuantity and Rate of Distinct FilesRate of change of popularity of filesOpen QuestionsReferencesMeasurements ofPeer-to-Peer SystemsPradnya KarbhariNov 25th, 2003CS 8803: Network Measurements SeminarIntroduction to Peer-to-Peer (P2P) systemsEnd-systems (or peers), are capable of behaving as clients and servers of data, hence system is scalable and reliablePeers participation is voluntary, membership is dynamic, hence topology keeps changingMost popularly used for file sharing, hence peer-to-peer systems have become synonymous with peer-to-peer file sharing networksClassification of P2P systemsP2P computation (e.g. seti@home)P2P communication (instant messaging)P2P file-sharing networksCentralized (e.g. Napster)DecentralizedStructured (e.g. Chord, CAN, Pastry, Tapestry)Unstructured (e.g. Gnutella, Kazaa, Freenet, eDonkey, eMule, Direct Connect, …)Popularity of unstructured decentralized P2P networksGnutella host count, maintained by Limewire (http://www.limewire.com)good scope for measurement studies because:deployed and widely useduse a lot of bandwidth during data transfer, hence a concern for network operatorsquite a few measurement studies have been done on these systems, some of which we will discuss in this seminarOutlineCharacterization of users of P2P systemsSaroiu, et.al., “A Measurement Study of Peer-to-Peer File Sharing Systems”, MMCN, 2002.Effect of P2P traffic on the underlying networkSen, et.al., “Analyzing peer-to-peer traffic across large networks”, IMW’02Peer-to-Peer TopologiesRipeanu, et.al., “Mapping the Gnutella Network: Properties of Large-Scale Peer-to-Peer Systems and Implications for System Design”, IEEE Internet Computing, 2002.Searching on the P2P networkSripanidkulchai, “The popularity of Gnutella queries and its implications on scalability”, 2001Deciphering proprietary P2P systems (like Kazaa)Leibowitz, et.al., “Deconstructing the Kazaa Network”, WIAPP, 2003.Gnutella protocol overviewConnecting to the Gnutella networkbootstrap using GWebCache system and locally cached hostlistPing/Pong messages are exchanged with potential neighborsSearching on the networkQuery messages are flooded on the networkQueryHit messages are received (back-propagated along Query path) from peers having the requested contentDownloading the contentpeers download files directly from peers having the requested contentCharacterization of Users of P2P systemsS. Saroiu, P. Gummadi and S. Gribble, “A Measurement Study of Peer-to-Peer File Sharing Systems”, MMCN’02.first paper to characterize p2p file sharing systemsGoal: To analyze the following user characteristicslatencylifetime of peersbottleneck bandwidthnumber of files shared and downloadeddegree of cooperationmethodology: active crawlingsystems studied: Napster and Gnutelladata collection: May 2001Saroiu et.al., “A Measurement Study of Peer-to-Peer File Sharing Systems”, MMCN, 2002Measurement Methodologyactive crawling of the Napster and Gnutella systemsNapster: issued queries for popular content, and then queried central server for peer informationGnutella: used ping/pong messages in protocol to get metadata about peers, and then their neighbors and so onparallel measurement for:peer lifetime- periodic probing of peers obtained from crawlersoffline if no response to TCP SYNinactive if response to TCP SYN is a TCP RSTactive if accepts the incoming TCP connection on that portlatency- RTT measurements from one hostbottleneck link bandwidth- active probing using Sprobe, a tool they developed based on packet-pair dispersion techniqueSaroiu et.al., “A Measurement Study of Peer-to-Peer File Sharing Systems”, MMCN, 2002Host Lifetime analysis20% peers in Napster, Gnutella have IP-level uptime of 93% or moreNapster peers have higher application uptimes than Gnutella peersthe best 20% of Napster peers have uptime of 83% or more and the best 20% of Gnutella peers have uptime of 45% or moremedian session duration is 60 minutes for Napster and GnutellaSaroiu et.al., “A Measurement Study of Peer-to-Peer File Sharing Systems”, MMCN, 2002Latency analysis (Gnutella)20% peers have a latency of at most 70ms and 20% have a latency of at least 280mscorrelation between downstream bottleneck bandwidth and latency: two clusters for modems (20-60Kbps, 100-1000ms) and broadband (1Mbps, 60-300ms)Saroiu et.al., “A Measurement Study of Peer-to-Peer File Sharing Systems”, MMCN, 2002Bottleneck Bandwidth Analysis (Gnutella)92% Gnutella peers have downstream bottleneck bandwidth of at least 100Kbps22% peers have upstream bottleneck bandwidth of 100Kbps or lesspeers are unsuitable to serve contentSaroiu et.al., “A Measurement Study of Peer-to-Peer File Sharing Systems”, MMCN, 2002Downloads, Uploads and Shared Filesrelative number of downloads and uploads varies significantly across bandwidth classesclear client/server behavior of different classesSaroiu et.al., “A Measurement Study of Peer-to-Peer File Sharing Systems”, MMCN, 2002Shared files v/s Shared Data(Napster and Gnutella)Strong correlation between number of files shared and amount of shared MB of dataslope of both lines is 3.7MB, the size of a typical MP3 audio