Interest-Based Self-Organizing Peer-to-Peer Networks: A Club Economics Approach

Abstract

Improving the information retrieval (IR) performance of peer-to-peer networks is an important and challenging problem. Recently, the computer science literature has attempted to address this problem by improving IR search algorithms. However, in peer-to-peer networks, IR performance is determined by both technology and user behavior, and very little attention has been paid in the literature to improving IR performance through incentives to change user behavior.We address this gap by combining the club goods economics literature and the IR literature to propose a next generation file sharing architecture. Using the popular Gnutella 0.6 architecture as context, we conceptualize a Gnutella ultrapeer and its local network of leaf nodes as a "club" (in economic terms). We specify an information retrieval-based utility model for a peer to determine which clubs to join, for a club to manage its membership, and for a club to determine to which other clubs they should connect.We simulate the performance of our model using a unique real-world dataset collected from the Gnutella 0.6 network. These simulations show that our club model accomplishes both performance goals. First, peers are self-organized into communities of interest -in our club model peers are 85% more likely to be able to obtain content from their local club than they are in the current Gnutella 0.6 architecture. Second, peers have increased incentives to share content -our model shows that peers who share can increase their recall performance by nearly five times over the performance offered to free-riders. We also show that the benefits provided by our club model outweigh the added protocol overhead imposed on the network for the most valuable peers.

« Pair à Pair » : les véritables enjeux économiques

“…Si le point commun des réseaux P2P repose sur le fait que le contenu y est toujours distribué, on peut distinguer différents types de réseaux (voir par exemple sur l'architecture de Gnutella, Asvanund et al, 2004). L'architecture du réseau peut en effet être : (i) complètement centralisée ; (ii) complètement décentralisée ; ou (iii) hiérarchique.…”

“…Les réseaux P2P peuvent à ce titre relever de l'analyse des marchés biface et financer par l'exploitation des données, via la publicité par exemple, le réseau lui-même. Asvanund et al (2004) se focalisent sur la meilleure manière d'améliorer la performance des réseaux P2P quant à la récupération des données. La performance dans la récupération des données est déterminée à la fois par la technologie et par le comportement des utilisateurs.…”

“…Pour résoudre ces difficultés, Asvanund et al (2004) développent un mécanisme de formation de communauté bien organisée afin d'augmenter la performance du réseau sans avoir à faire appel à un planificateur centralisé ou à du contenu identifié de manière unique. Toutefois, ils ne prennent pas en compte les possibles enjeux commerciaux de la récupération de données ni la valeur créée par les réseaux P2P du fait de la récupération de ces données.…”

Clustering and Similarity Analysis of File Names in EDonkey Systems

“…Supposing users' interest is stable and the file category shared by one user can represent his/her own interest, Asvanund A [2] proposed to classify all users into different interest group by calculating the frequency vector's Jensen-Shannon differentiation value of shared file names. In later queries, those which are in the same group will be firstly queried, because these users is the most probable candidates to return intended resources.In fact, Crespo A [3] and Singh S [4] have also raised similar ideas to group users.…”

An Association Rule Based Grid Resource Discovery Method

“…Currently, the ontology based resource discovery methods rises which abstract resource logic description from user's behavior model and resolve the problem brought by keyword-based information organization and retrieval [4]. Through analyzing those existing resource discovery methods in grid, we can deduce that resource in grid is lack of complete representation since resource description language is not unified in heterogeneous system and the process of resource discovery neglects user's behavior or interest while some researches show that the user's interaction behavior or interest is stable relatively [5], [6]. In order to make it possible to obtain what the users really need quickly and accurately from the resource query results, it becomes more meaningful to research on resource discovery method based on semantics and interaction behavior from resources scheduling processing.…”