Skip to main content
Presentation
Multi-tracker collaboration in bittorrent systems
International Conference on Software Engineering and Knowledge Engineering (2010)
  • Sonia Gulrajani, swan jose state university
  • Anuja Oka, San José State University
  • Xiao Su, San Jose State University
Abstract
While BitTorrent has fast gained popularity and is being extensively used to share files among users, its performance is limited by the fact that independent torrent networks for the same file are unable to collaborate. This can result in starved torrent networks, due to the incognizance of the existing active torrent networks for the same files. One way to address this problem is to have the trackers of such torrent networks collaborate with each other by exchanging peer lists, thereby resulting in an extended peer pool. Along with helping the starved networks, multi-tracker collaboration can also enhance torrent performance, with respect to download speed by bringing together independent healthy torrent networks for the same files and creating a virtual common torrent network. In this paper we propose to introduce multi-tracker collaboration into BitTorrent protocol, and study its performance gain by implementing it in the popular BNBT tracker software.
Keywords
  • Collaborative distribution,
  • File sharing,
  • Incentives,
  • Peer-to-peer
Disciplines
Publication Date
July, 2010
Location
Redwood City, CA; United States
Comments
July 1-3, 2010
Citation Information
Sonia Gulrajani, Anuja Oka and Xiao Su. "Multi-tracker collaboration in bittorrent systems" International Conference on Software Engineering and Knowledge Engineering (2010)
Available at: http://works.bepress.com/xiao_su/25/