Unraveling BitTorrent's File Unavailability: Measurements and Analysis

Abstract

BitTorrent suffers from one fundamental problem: the long-term availability of content. This occurs on a massivescale with 38% of torrents becoming unavailable within the first month. In this paper we explore this problem by performing two large-scale measurement studies including 46K torrents and 29M users. The studies go significantly beyond any previous work by combining per-node, per-torrent and system-wide observations to ascertain the causes, characteristics and repercussions of file unavailability. The study confirms the conclusion from previous works that seeders have a significant impact on both performance and availability. However, we also present some crucial new findings: (i) the presence of seeders is not the sole factor involved in file availability, (ii) 23.5% of nodes that operate in seedless torrents can finish their downloads, and (iii) BitTorrent availability is discontinuous, operating in cycles of temporary unavailability. Due to our new findings, we consider it is important to revisit the solution space; to this end, we perform large-scale trace-based simulations to explore the potential of two abstract approaches.

Characterizing the file hosting ecosystem: A view from the edge

“…The difference is that in case of P2P, a file may persist for a long time, but it may not have enough peers for the content to be downloaded fully. Kaune et al found that a lack of seeders often resulted in file unavailability (86% of the time) [18]. Private torrent sites institute a minimum upload-to-download ratio to increase file availability, however, these schemes have been shown to be subject to collusion [22].…”

A mathematical model for file fragment diffusion and a neural predictor to manage priority queues over BitTorrent

“…One of the mechanisms proposed to increase file availability has been to use a multi-torrent, i.e., for ensuring fairness, instead of forcing users to stay longer, contribution is provided to uploaders for fragments belonging to different files (Guo et al, 2005). Similarly, Kaune et al (2010) show that, by using the multi-torrent, availability can be easily increased, and confirm that fast replication of rare fragments is essential. Furthermore, bundling, i.e., the dissemination of a number of related files together, has been proposed to increase availability (Menasche et al, 2009).…”

“…Fragments availability is measured by the number of peers storing a file fragment at a given moment, and periodically computed by a tracker server storing peer ids, fragments held, and files requested (Cohen, 2003). In order to compute the distribution priority for each file fragment, it is of paramount importance to synchronize with the tracker and to get frequent updates since the status of the BitTorrent network is prone to rapid changes, due to the high variability of the number and availability of peers that can leave the system at any time (Kaune et al, 2010). This occurs so frequently that such a fundamental * Corresponding author BitTorrent mechanism may become ineffective, and as a result some fragments can quickly become unavailable.…”

A Task-Based Model for the Lifespan of Peer-to-Peer Swarms

“…Kaune et al also focused on the availability in modern systems. They performed widespread measurements and found that seeders have a significant impact on swarm availability [10]. Then they tried different incentives to improve the availability and gave a comparison study on them [11].…”