A dynamic interoperability model for an emergent middleware framework

• Main Author: Nundloll, Vatsala
• Other Authors: Blair, Gordon
• Published: Lancaster University 2013
• Online Access: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.755062

The rapid changing world of computing has sparked off a major increase in the complexity, heterogeneity and dynamicity of distributed systems. Consequently, standard middleware platforms are unable to cope with the extreme heterogeneity and dynamicity of this new generation of distributed systems. Furthermore, given new trends in mobile/pervasive applications, distributed systems are required to connect to one another at run time, implying that heterogeneities arising in systems need to be resolved on the fly. This ability of a system to interact with a different system is known as interoperability. Existing middleware interoperability solutions cannot deal with dynamic interoperability because of their static and hand-crafted nature. Hence, more advanced solutions that exceed the state-of-the-art in middleware, are required to handle interoperability on the fly. This thesis investigates the challenges of dynamic interoperability and how to devise an emergent middleware to enable such dynamic interoperation. To overcome the heterogeneities arising at runtime, the thesis also investigates the approach of the Semantic Web community to employ semantic reasoning of concepts at the application level. The thesis maintains that one such notable contribution of this community, the use of ontologies, has proved to play a significant role in the set up of such an emergent middleware framework. As a result, the thesis proposes a framework with 3 distinct design principles - matching, classifying and mapping – to tackle dynamic interoperability at the message level of systems, and also highlights the cross-cutting role played by ontologies in the framework. The experimental evaluation of the framework shows the framework is able to tackle the heterogeneity arising in messages at runtime, and also highlights the significance of linguistic techniques in assisting ontologies at the matching stage. Finally, the performance evaluation denotes how the framework behaves at runtime and justifies how the framework performs its intended purpose.