Measurements, analysis and modeling of private tracker sites

Xiaowei Chen*, Xiaowen Chu, Yixin Jiang, Fengyuan Ren

*Corresponding author for this work

Research output: Chapter in book/report/conference proceedingConference proceedingpeer-review

3 Citations (Scopus)

Abstract

BitTorrent plays a very important role in the current Internet content distribution. When BitTorrent public tracker sites are suffering from free-riding problem, private tracker sites (PTs) work very well because of Share Ratio Enforcement (SRE) which is an auxiliary effective incentive mechanism. Understanding PTs is essential to content distribution. We have crawled and traced 15 tracker sites with over 3.5 million torrents for 7 months. We first provide taxonomy of PTs, and then present measurement study on the characteristics of PTs from the user viscosity, torrents evolution, user behaviors, and content distribution. Some of the features are apparently different from public trackers. Furthermore, we analyze SRE mechanism and auxiliary credit system, and use game theory to study effectiveness of SRE mechanism. There exists "uploading starvation" phenomenon in private trackers. We model SRE mechanism and propose an improved SRE mechanism to further incent the users and enhance the performance of private trackers.

Original languageEnglish
Title of host publication2010 IEEE 18th International Workshop on Quality of Service, IWQoS 2010
DOIs
Publication statusPublished - 2010
Event2010 IEEE 18th International Workshop on Quality of Service, IWQoS 2010 - Beijing, China
Duration: 16 Jun 201018 Jun 2010

Publication series

NameIEEE International Workshop on Quality of Service, IWQoS
ISSN (Print)1548-615X

Conference

Conference2010 IEEE 18th International Workshop on Quality of Service, IWQoS 2010
Country/TerritoryChina
CityBeijing
Period16/06/1018/06/10

User-Defined Keywords

  • BitTorrent
  • Content distribution
  • Incentive mechanism
  • Peer-to-peer networks
  • Private tracker

Fingerprint

Dive into the research topics of 'Measurements, analysis and modeling of private tracker sites'. Together they form a unique fingerprint.

Cite this