A Hierarchical Semantic Overlay for P2P Search

TL;DR

This paper introduces a hierarchical semantic overlay network for efficient P2P data search, combining structured and unstructured overlays to improve search performance over wide-area networks.

Contribution

It proposes a novel hierarchical semantic overlay architecture that organizes peers into semantic clusters and overlays, enhancing search efficiency and cost-effectiveness.

Findings

Improved search efficiency through combined overlay strategies

Reduced search and maintenance costs

Effective handling of heterogeneous data in P2P networks

Abstract

In this paper, we propose a hierarchical semantic overlay network for searching heterogeneous data over wide-area networks. In this system, data are represented as RDF triples based on ontologies. Peers that have the same semantics are organized into a semantic cluster, and the semantic clusters are self-organized into a one-dimensional ring space to form the toplevel semantic overlay network. Each semantic cluster has its low-level overlay network which can be built using an unstructured overlay or a DHT-based overlay. A search is first forwarded to the appropriate semantic cluster, and then routed to the specific peers that hold the relevant data using a parallel flooding algorithm or a DHT-based routing algorithm. By combining the advantages of both unstructured and structured overlay networks, we are able to achieve a better tradeoff in terms of search efficiency, search cost and overlay maintenance cost.