BTWorld: towards observing the global BitTorrent file-sharing network

  • Authors:
  • Maciej Wojciechowski;Mihai Capotă;Johan Pouwelse;Alexandru Iosup

  • Affiliations:
  • TU Delft, The Netherlands;TU Delft, The Netherlands;TU Delft, The Netherlands;TU Delft, The Netherlands

  • Venue:
  • Proceedings of the 19th ACM International Symposium on High Performance Distributed Computing
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

Today, the BitTorrent Peer-to-Peer file-sharing network is one of the largest Internet applications---it generates massive traffic volumes, it is deployed in thousands of independent communities, and it serves millions of unique users worldwide. Despite a large number of empirical and theoretical studies, observing the state of the global BitTorrent network remains a grand challenge for the BitTorrent community. To address this challenge, in this work we introduce BT-World, an architecture for observing the global BitTorrent network without help from the ISPs. We design BTWorld around three main features specific to BitTorrent measurements. First, our architecture is able to find public trackers, that is, the BitTorrent components that offer unrestricted service to peers around the world. Second, by observing the state of these trackers, BTWorld obtains information about the performance, scalability, and reliability of BitTorrent. Third, BTWorld is designed to pre-process the large volumes of recorded data for later analysis. We demonstrate the viability of our architecture by deploying it in practice, to observe and analyze one week of operation of a large part of the global BitTorrent network--over 10 million swarms and tens of millions of concurrent users. We also show that BT-World can shed light on BitTorrent phenomena, such as the presence of spam trackers and giant swarms.