BitTorrent's dilemma: Enhancing reciprocity or reducing inequity

Abstract

Abstract-Enhancing reciprocity has been one of the primary motivations for the design of incentive policies in BitTorrent-like P2P systems. Reciprocity implies that peers need to contribute their bandwidth to other peers if they want to receive bandwidth in return. However, the over-provisioning that characterizes today's BitTorrent communities and the development of many next-generation P2P systems with real-time constraints (e.g., for live and on-demand streaming) suggest that more effort can be devoted to reducing the inequity (i.e., the difference of service received) among peers, rather than only enhancing reciprocity. Inspired by this observation, in this work we analyze in detail several incentive mechanisms that are used in BitTorrent systems, and explore several strategies that influence the balance between reciprocity and equity. Our study shows that (i) reducing inequity leads to a better overall system performance, and (ii) the behavior of seeders (i.e., peers that hold a complete copy of the file and upload it for free) influences whether reciprocity is enhanced or inequity reduced.

Fast download but eternal seeding: The reward and punishment of Sharing Ratio Enforcement

“…Taking inspiration from this concept, we propose the third strategy which takes into account the effort of users in terms of their seeding times. Previous studies have shown that the effort-based incentive policy applied in the leeching process improves the systemwide performance [20], [14]. We expect the same improvement when this effort-based methodology is applied in a private community.…”

“…We assume that in an undersupplied swarm both leechers and seeders can fully utilize their upload capacities, i.e., u i = U i . This assumption has been validated by previous studies [17], [14]. In this situation, within a time interval ∆ the total upload volume that can be provided by all peers should be no larger than the total download volume required by all…”

“…We follow a similar fluid modeling approach as in [18], [17], [14] and extend it by including SRE and considering multiple user behaviors derived from real world observations. The notation we use is shown in Table I.…”

“…In this way, peers are roughly clustered according to their capacities, and peers with similar capacities have similar performance [17], [14]. Let x i (t) and y i (t) represent the number of leechers and lazy-seeding seeders in class i at a particular time t, and let T i represent the average seeding time of lazy-seeding peers in class i, then the evolution of x i (t) and y i (t) can be described as:…”

“…With x i leechers, y i lazy-seeding seeders, and s i overseeding seeders in class i in a steady state, the average downloading speed of a peer in class i can be calculated by solving the system of equations proposed in our previous work [14]:…”