Metadata-Based Architectures for Information Brokering

Summary

A metadata-based architecture for information brokering was described in this chapter. We identified and discussed two basic components of this architecture, metadata and vocabulary. These components were described in greater detail by describing their sub-components and their respective functionalities. The contributions made by each of the components towards tackling information overload were also discussed. A set of “SEA” properties, scalability, extensibility, and adaptability that brokering systems should have, were identified and defined. Issues relating to the above properties as relating to the different components of the architecture were also discussed.

We also discussed the evolution of architectures to support interoperability. Reference architectures for federated multidatabase systems and mediatorbased systems were compared to the brokering architecture. The “SEA” properties of these architectures were also discussed. We concluded by presenting a hybrid agent based brokering architecture for information brokering.