Implementing Database Coordination in P2P Networks

We are interested in the interaction of databases in Peer-to-Peer (P2P) networks. In this paper we propose a new solution for P2P databases, that we call database coordination. We see coordination as managing semantic interdependencies among databases at runtime. We propose a data coordination model where the notions of Interest Groups and Acquaintances play the most crucial role. Interest groups support the formation of peers according to data models they have in common; and acquaintances allow for peers inter-operation. Finally, we present an architecture supporting database coordination and show how it is implemented on top of JXTA.