Adaptive sharing of large resources in P2P networks

ABSTRACT

A peer-to-peer(P2P) system comprises a network of nodes that are capable of sharing and exchanging resources with one another. Recent studies of P2P networks show that many resources exchanged between users are considerably large files that require significant download times, consume the majority of the network bandwidth, and also occupy substantial storage space on the node providing the resource. In such a scenario, it would be inefficient for a node to store a large resource that is rarely, or never requested by other nodes, or, to share a large resource with a node that is already acquiring the resource from another source. These inefficiences can be mitigated if a node dynamically determines and updates its decision to store and share large resources. Here, we describe an agent enabled adaptive strategy for a node to share large resources based on the expected availability of the resource in the network. Experimental results of our adaptive sharing strategy show that a saving of 8--12 downloads/resource, accounting for 50--70 MB of saved data transfer/resource, can be achieved without performance deterioration in the P2P network.

REFERENCES

• 1: E. Adar and B. Huberman. "Free Riding on Gnutella," First Monday.
• 2: Sviatoslav Braynov , Tuomas Sandholm, Incentive compatible mechanism for trust revelation, Proceedings of the first international joint conference on Autonomous agents and multiagent systems: part 1, July 15-19, 2002, Bologna, Italy [doi>10.1145/544741.544814]
• 3: Luís Henrique M. K. Costa , Marcelo Dias De Amorim , Serge Fdida, Reducing latency and overhead of route repair with controlled flooding, Wireless Networks, v.10 n.4, p.347-358, July 2004 [doi>10.1023/B:WINE.0000028539.73976.64]
• 4: Mark E. Crovella , Murad S. Taqqu , Azer Bestavros, Heavy-tailed probability distributions in the World Wide Web, A practical guide to heavy tails: statistical techniques and applications, Birkhauser Boston Inc., Cambridge, MA, 1998
• 5: Gnutella, URL http://www.gnutella.com
• 6: Krishna P. Gummadi , Richard J. Dunn , Stefan Saroiu , Steven D. Gribble , Henry M. Levy , John Zahorjan, Measurement, modeling, and analysis of a peer-to-peer file-sharing workload, Proceedings of the nineteenth ACM symposium on Operating systems principles, October 19-22, 2003, Bolton Landing, NY, USA
• 7: Philippe Golle , Kevin Leyton-Brown , Ilya Mironov, Incentives for sharing in peer-to-peer networks, Proceedings of the 3rd ACM conference on Electronic Commerce, p.264-267, October 14-17, 2001, Tampa, Florida, USA [doi>10.1145/501158.501193]
• 8: A. Mas-Colell, M. Whinston and J. Green, "Microeconomic Theory," Oxford University Press, 1995.
• 9: L. Massoulie, "Stability of Distributed Congestion Control with Heterogeneous Feedback Delays," Microsoft Research Technical Report, 2000.
• 10: Mojo Nation, URL http://sourceforge.net/projects/mojonation
• 11: The FastTrack Protocol URL http://cvs.berlios.de
• 12: Sylvia Ratnasamy , Paul Francis , Mark Handley , Richard Karp , Scott Schenker, A scalable content-addressable network, Proceedings of the 2001 conference on Applications, technologies, architectures, and protocols for computer communications, p.161-172, August 2001, San Diego, California, United States
• 13: Ion Stoica , Robert Morris , David Liben-Nowell , David R. Karger , M. Frans Kaashoek , Frank Dabek , Hari Balakrishnan, Chord: a scalable peer-to-peer lookup protocol for internet applications, IEEE/ACM Transactions on Networking (TON), v.11 n.1, p.17-32, February 2003 [doi>10.1109/TNET.2002.808407]
• 14: Bin Yu , Munindar P. Singh, Searching social networks, Proceedings of the second international joint conference on Autonomous agents and multiagent systems, July 14-18, 2003, Melbourne, Australia [doi>10.1145/860575.860587]

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.2 ARTIFICIAL INTELLIGENCE

Additional Classification:
  I. Computing Methodologies
  I.2 ARTIFICIAL INTELLIGENCE
        I.2.11 Distributed Artificial Intelligence

General Terms:
Algorithms, Economics, Management

Keywords:
peer-to-peer networks, probabilistic sharing, resource management, revelation mechanism