On Analyzing Free-Riding Behavior in BitTorrent Communities
Abhijit Das, Arup Kumar Bhattacharjee
Abstract
Abhijit Das, Arup Kumar Bhattacharjee
Abstract
BitTorrent is undoubtedly the most popular P2P file sharing systems in existence that has been able to attract millions of users. BitTorrent depends on global components for file search, engages a moderator system to guarantee the data integrity of file, and uses an exchange mechanism for file sharing in order to discourage users from free-riding. We conduct one of the largest measurement and analysis of 4 BitTorrent communities in 2 different time periods to understand free-riding behavior. Although it is well-known that BitTorrent is exposed to free-riders for optimistic unchoking, this paper demonstrates that sharing communities are easily cheated by free-riders and are particularly appealing to them. We have observed that the proportion of seeders are consistently higher across BitTorrent communities, thereby encouraging free-riding. Contrary to the popular belief, our analysis reveals that free-riders does not deteriorates the system performance. We also report some unusual peer behaviors existing in BitTorrent communities.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
BitTorrent is undoubtedly the most popular P2P file sharing systems in existence that has been able to attract millions of users. BitTorrent depends on global components for file search, engages a moderator system to guarantee the data integrity of file, and uses an exchange mechanism for file sharing in order to discourage users from free-riding. We conduct one of the largest measurement and analysis of 4 BitTorrent communities in 2 different time periods to understand free-riding behavior. Although it is well-known that BitTorrent is exposed to free-riders for optimistic unchoking, this paper demonstrates that sharing communities are easily cheated by free-riders and are particularly appealing to them. We have observed that the proportion of seeders are consistently higher across BitTorrent communities, thereby encouraging free-riding. Contrary to the popular belief, our analysis reveals that free-riders does not deteriorates the system performance. We also report some unusual peer behaviors existing in BitTorrent communities.
Key concepts: BitTorrent, Free riding, Computer science, BitTorrent tracker, Free rider problem, File sharing, Internet privacy, Peer-to-peer