Adaptive Distributed b-Matching in Overlays with Preferences
Paper i proceeding, 2012

An important function of overlay networks is the facilitation of connection, interaction and resource sharing between peers. The peers may maintain some private notion of how a “desirable” peer should look like and they share their bounded resources with peers that they prefer better than others. Recent research proposed that this problem can be modeled and studied analytically as a many-to-many matching problem with preferences. The solutions suggested by the latter proposal guarantee both algorithmic convergence and stabilization, however they address static networks with specific properties, where no node joining or leaving is considered. In this paper we present an adaptive, distributed algorithm for the many-to-many matching problem with preferences that works over any network, provides a guaranteed approximation for the total satisfaction in the network and guarantees convergence. In addition, we provide a detailed experimental study of the algorithm that focuses on the levels of achieved satisfaction as well as convergence and reconvergence speed. Finally, we improve, both for static and dynamic networks, the previous known approximation ratio.

Författare

Georgios Georgiadis

Chalmers, Data- och informationsteknik, Nätverk och system

Marina Papatriantafilou

Chalmers, Data- och informationsteknik, Nätverk och system

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

03029743 (ISSN) 16113349 (eISSN)

Vol. 7276 208-223
978-3-642-30849-9 (ISBN)

Styrkeområden

Informations- och kommunikationsteknik

Ämneskategorier

Datavetenskap (datalogi)

DOI

10.1007/978-3-642-30850-5_19

ISBN

978-3-642-30849-9

Mer information

Skapat

2017-10-07