Agents and Peer-to-Peer Computing: Towards P2P-Based Resource Allocation in Competitive Environments

Peer-to-peer frameworks are known to be robust and scalable to large numbers of agents. Recent resource allocation studies have leveraged this by using peer-to-peer frameworks for the implementation of resource matching algorithms. In this paper, we present a matching protocol for multiagent resource allocation in a competitive peer-to-peer environment; this work marks the first solution to the resource matching problem in this type of environment. Our approach makes use of micropayment techniques, along with concepts from random graph theory and game theory. We provide an analytical characterization of our protocol, and specify how an agent should choose optimal values for the protocol