On Using Matching Theory to Understand P2P Network Design

Dmitry Lebedev (1) , Fabien Mathieu (1) , Laurent Viennot (2) , Anh-Tuan Gai (2) , Julien Reynier (3) , Fabien de Montgolfier (2, 4)

1 FT R&D - France Télécom Recherche & Développement
2 GANG - Networks, Graphs and Algorithms
3 LIENS - Laboratoire d'informatique de l'école normale supérieure
4 LIAFA - Laboratoire d'informatique Algorithmique : Fondements et Applications

This paper aims at giving insight on the stability of collaboration choices in P2P networks. We study networks where exchanges between nodes are driven by the will of getting the best service available. This is the case for most existing P2P networks. We explore an evolution model derived from stable roommates theory that accounts for heterogeneity between nodes. We show that most P2P applications can be modeled using stable matching theory. This is the case whenever preference lists can be deduced from the exchange policy. In many cases, the preferences lists are characterized by an interesting acyclic property. This leads us to the notion of acyclic preferences. We show that P2P networks with acyclic preferences possess a unique stable state with good convergence properties.

Mots clés: P2P networks; stable matchings; convergence properties

Domaines: Algorithme et structure de données [cs.DS]; Informatique et théorie des jeux [cs.GT]

Dmitry Lebedev, Fabien Mathieu, Laurent Viennot, Anh-Tuan Gai, Julien Reynier, et al.. On Using Matching Theory to Understand P2P Network Design. INOC 2007, International Network Optimization Conference, 2007, Spa, Belgium. ⟨hal-00159678⟩