4.5 Article

SocialTrust: Enabling long-term social cooperation in peer-to-peer services

Journal

PEER-TO-PEER NETWORKING AND APPLICATIONS
Volume 7, Issue 4, Pages 525-538

Publisher

SPRINGER
DOI: 10.1007/s12083-013-0198-2

Keywords

Incentive mechanism; Peer-to-peer networks; Sybil-proof; Small world

Ask authors/readers for more resources

Peer-to-peer(P2P) services heavily rely on users' cooperation to achieve desired performance. However, most current P2P systems only encourage short-term and direct cooperation between peers. The lack of incentives for long term and indirect cooperation has severely limited the performance of P2P systems. On the other hand, recent measurements on large-scale networks show that peers' behavior often demonstrates strong social patterns. In this paper, we design and implement a social P2P network, named SocialTrust, based on peers' common interests. In SocialTrust, each peer tries to find a small number of friends and maintains long term social links with them. We also propose a distributed trust mechanism. The trust between two friends reflects their cooperation level and serves as the credit limit between them. A peer with higher trust can download data from its friends more efficiently. The trust can be propagated among friends to support indirect reciprocity. We formally prove that the proposed distributed trust mechanism is secure and can defend against various forms of attacks. By adding asmall number of long term social links to the existing P2P network, SocialTrust relaxes the constraint of direct incentive mechanisms and encourages peers to perform various forms of long-term cooperation. Both trace-driven simulation and real Internet experiments show that SocialTrust can significantly improve file availability and download performance of current P2P file sharing systems.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.5
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available