Were You There? Bridging the Gap to Unveil Users' Online Sessions in Networked, Distributed Systems

The comprehension of users' behavior is paramount for evaluating improvements to networked, distributed systems. To this end, several strategies have been proposed to obtain traces, based on the capture of usage information, which can then serve for evaluation purposes. One main strategy consists of taking snapshots of online users, using instrumented clients. In spite of its popularity, related proposals have fallen short in ensuring accuracy of obtained data. For a variety of reasons, users may fail to appear in some snapshots, although online. In this paper, we propose a methodology to correct ill-collected snapshots and build more accurate traces from them. In summary, we estimate the probability that a given snapshot is missing some users. The snapshot is corrected if the probability exceeds a given threshold. We use ground-truth data to assess the effectiveness of our methodology. The impact of our proposal is evidenced by means of an analysis of traces from a BitTorrent swarm.

[1]  Mikel Izal,et al.  Dissecting BitTorrent: Five Months in a Torrent's Lifetime , 2004, PAM.

[2]  Jean-Chrysostome Bolot,et al.  Location patterns of mobile users: A large-scale tudy , 2013, 2013 Proceedings IEEE INFOCOM.

[3]  Xiaoning Ding,et al.  A performance study of BitTorrent-like peer-to-peer systems , 2007, IEEE Journal on Selected Areas in Communications.

[4]  Taoufik En-Najjary,et al.  Long Term Study of Peer Behavior in the kad DHT , 2009, IEEE/ACM Transactions on Networking.

[5]  Reza Rejaie,et al.  Is content publishing in BitTorrent altruistic or profit-driven? , 2010, CoNEXT.

[6]  Diane E. Vaughan,et al.  A Survey of the Coupon Collector’s Problem with Random Sample Sizes , 2007 .

[7]  Maximilian Michel,et al.  Characterization of BitTorrent swarms and their distribution in the Internet , 2011, Comput. Networks.

[8]  Luciano Paschoal Gaspary,et al.  Segurança em redes P2P : princípios, tecnologias e desafios , 2006 .

[9]  Di Wu,et al.  Unraveling the BitTorrent Ecosystem , 2011, IEEE Transactions on Parallel and Distributed Systems.

[10]  James F. Kurose,et al.  On the steady-state of cache networks , 2013, 2013 Proceedings IEEE INFOCOM.

[11]  Xiaoning Ding,et al.  Measurements, analysis, and modeling of BitTorrent-like systems , 2005, IMC '05.

[12]  Jussi Kangasharju,et al.  Measuring large-scale distributed systems: case of BitTorrent Mainline DHT , 2013, IEEE P2P 2013 Proceedings.

[13]  Walid Dabbous,et al.  Spying the World from Your Laptop: Identifying and Profiling Content Providers and Big Downloaders in BitTorrent , 2010, LEET.

[14]  Akihiro Nakao,et al.  Measuring BitTorrent swarms beyond reach , 2011, 2011 IEEE International Conference on Peer-to-Peer Computing.

[15]  Akihiro Nakao,et al.  A Resource-Efficient Method for Crawling Swarm Information in Multiple BitTorrent Networks , 2011, 2011 Tenth International Symposium on Autonomous Decentralized Systems.