Modelling Peer-to-Peer Data Networks Under Complex System Theory

Abstract

A Peer-to-peer Data Network (PDN) is an open and evolving society of peer nodes that assemble into a network to pool and share their data (or more generally, their resources represented by data) for mutual benefit. By an interesting analogy to a democratic human society, when nodes join the PDN society, while they agree to follow a restricted set of common rules in interaction with their peers (i.e., the social rules governing the PDN society), they preserve their autonomy as individuals. For example, as part of their social obligations all PDN nodes (or at least those who are good PDN citizens) create and maintain connection with a set of neighbor nodes and participate in cooperative query processing (e.g., forwarding search queries for data discovery). Aside from the social rules, the PDN leaves the behavior of the individual nodes unregulated and flexible, to be managed by their users based on their individual preferences and/or to allow for natural uncertainties and constraints. For instance, nodes may join and leave the PDN society as they decide (by user decision or due to unwanted node/link failure), they control their own resources, and they select their neighbors according to their own administrative policy or physical constraints (e.g., connecting to the nodes that are both accessible and physically close as neighbors). In this sense, individual nodes are self-governed, autonomous, and independent. There is a trade-off between the extent of the social rules and the autonomy of the individual PDN nodes; the more extensive and interfering the social rules, the autonomy of the nodes is more restricted.