SOON: A scalable self-organized overlay network for distributed information retrieval

Abstract

Locating desirable resources and information from a large-scale distributed system such as P2P system and grid is a very important issue. However, the distributed, heterogeneous, and unstructured nature of the system makes this issue very challenging. In this paper, we propose Self-Organized Overlay Network (SOON), an unstructured P2P overlay architecture, to facilitate sharing and searching semantically heterogeneous contents. In particular, we have proposed a semantics-aware topology construction method to group nodes sharing similar semantics together to create small-worlds. For this purpose, we have designed an algorithm to extract a node's ontology summary and use that summary to compute the semantic similarity between nodes. With this semantic similarity defined, nodes are grouped accordingly, forming semantic virtual domains and clusters. Resource information integration and searching can be efficiently performed on top of this topology. © 2008 Springer-Verlag.