Improving robustness and fairness on BitTorrent-Like P2P systems

Author

Fenglin Qin ; Ju Liu ; Lina Zheng

Author_Institution

Sch. of Inf. Sci. & Eng., Shandong Univ., Jinan, China

fYear

2009

fDate

26-28 Aug. 2009

Firstpage

1

Lastpage

5

Abstract

Recent studies have shown that BitTorrent, one of the most popular peer-to-peer file-sharing systems, is confronted with two serious problems: robustness and fairness. In this paper, a partially decentralized trust management system for BitTorrent is presented to deal with the two problems. By extending and modifying the functionality of the tracker in BitTorrent swarm, the extended tracker serves as an authoritative trust computation agent (TCA) to calculate and disseminate global trust values for all peers, which complement with local trust values to identify and isolate malicious and selfish peers. Simulations show that the proposed trust management system significantly decreases the impact of malicious peers on the swarm performance, and also achieves a much fairer block allocation among peers in heterogeneous scenarios.

Keywords

fault tolerant computing; peer-to-peer computing; security of data; BitTorrent swarm; BitTorrent-like P2P systems; P2P system fairness; P2P system robustness; extended tracker; peer-to-peer file sharing systems; trust computation agent; Bandwidth; Computational modeling; Computer hacking; Degradation; Information science; Internet; Peer to peer computing; Protocols; Robustness; Thin film transistors;

fLanguage

English

Publisher

ieee

Conference_Titel

Communications and Networking in China, 2009. ChinaCOM 2009. Fourth International Conference on

Conference_Location

Xian

Print_ISBN

978-1-4244-4337-6

Electronic_ISBN

978-1-4244-4337-6

Type

conf

DOI

10.1109/CHINACOM.2009.5339915

Filename

5339915