||Enhancing Flexibility in Service Oriented Enterprise Application Integration through Semantic Web Technologies
||Baumeister, Hubert (Software Engineering, Department of Informatics and Mathematical Modeling, Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark)
||Technical University of Denmark, DTU, DK-2800 Kgs. Lyngby, Denmark
||As business processes become more and more involved with computer- and data-operations, interdependencies arise. Business processes are designed around the IT-infrastructure and applications and hardware solutions are modified to satisfy business processes. The domain of enterprise application integration (EAI) seeks to enable an organization to integrate its existing applications and to incorporate new applications and technologies into an existing IT infrastructure.
Due to the complexities of most EAI scenarios, modifications to an existing solution may result in a new solution which, while syntactically correct, may differ enough on the semantic level that data or sub-processes become nonsensical for existing processes and applications. While existing technologies serve their function very well on the syntactic level, the problem remains on the semantic level.
The thesis poses the following question:
Can semantic web technologies be used to enhance the flexibility of service oriented enterprise application integration?
The analysis performed in the thesis is based on an existing EAI framework. This framework works in practice, but challenges are faced when new systems or data are added to the integrated processes. Based on the analysis performed in this thesis, a new prototype framework is developed.
It was found that representing the integrated data domain as ontologies provide a common vocabulary and understanding of data across (and beyond) the enterprise. This reduces risks by providing a degree of transparency in an otherwise hard to understand data model. More importantly, many of the high-risk manual tasks such as service- and data-mapping and message routing may be automated by allowing the integration middleware to reason on the data it transmits. Automatic reasoning is made possible primarily through the incorporation of semantically enriched data.
To summarize, it was found that using semantics in EAI scenarios provides multiple benefits and significantly increases flexibility and the ability to adapt.
||Technical University of Denmark (DTU) : Kgs. Lyngby, Denmark
Creation date: 2009-10-01
Update date: 2009-11-04