A Hierarchical Architecture for Distributed EPCglobal Discovery Services

Efficient and scalable information discovery is one of the most important services in any large- scale Internet of Things (IoT) application, particularly in the EPCglobal Network. Although a number of distributed architectures have been proposed in the literature, both their scalability and their lookup time efficiency remain vulnerable, mainly because of their reliance on flat Peer-to-Peer (P2P) Networking. The purpose of this paper is to introduce a hierarchical distributed architecture for EPCglobal Discovery Services, called HEDSA, which improves the scalability and the lookup time of the flat P2P architectures, represented by FEDSA. The idea behind the hierarchy concept of HEDSA is that any Electronic Product Code (EPC) can be mapped to one and only one country, which is the issuing country of the corresponding company prefix. An emulation of FEDSA and HEDSA has been implemented on Planetlab using Chord algorithm, the objective being to compare the scalability and the lookup time of the two architectures. Several experiments have shown that HEDSA is much more efficient, both in terms of the number of hops and the lookup time, than FEDSA. Therefore, HEDSA is more suitable for large-scale IoT discovery services applications, such as the EPCglobal Network, provided that the identifiers can be mapped to one and only one geographical location.