Article in Proceedings INPROC-2003-22

BibliographyMariucci, Marcello; Dorda, Clemens; Mitschang, Bernhard: Design and Implementation of a Model-driven Earth Observation Integration Framework.
In: Kotsis, Gabriele (ed.); Bressan, Stephane (ed.); Catania, Barbara (ed.); Ibrahim, Ismail Khalil (ed.): Proceedings of the Fifth International Conference on Information Integration and Web-based Applications & Services: iiWAS '03; Jakarta, Indonesia, September 15-17, 2003. Vol. 170.
University of Stuttgart, Faculty of Computer Science, Electrical Engineering, and Information Technology.
pp. 215-225, english.
Oesterreichische Computer Gesellschaft, September 2003.
ISBN: 3-85403-170-10.
Article in Proceedings (Conference Paper).
CR-SchemaH.3.5 (Online Information Services)
J.2 (Physical Sciences and Engineering)
D.2.2 (Software Engineering Design Tools and Techniques)
KeywordsModel-driven Architecture, Process-oriented Application Integration, Model-driven Integration Process, Repository System
Abstract

An Earth Observation (EO) integration framework is an application integration solution for supporting the development and execution of EO services. EO services are based on the intensive use of large data sets from space, and require the tight cooperation of several distributed experts. They can be characterized as highly flexible structures, which constantly need to be adapted to evolving spacecraft and processing technologies. In this paper we introduce a model-driven approach for an EO integration framework. We describe a comprehensive integration model that adequately copes with the flexible development, customization, and execution of EO services. The accurate treatment of related model instances throughout the software life cycle significantly enhances the EO service development process in terms of quality, reuse, and adaptability. We discuss technological aspects for the realization of such an integration framework, and outline our prototype implementation that is mainly built upon commercial products. We demonstrate that such a model-driven approach can be realized by employing repository technology for managing model related issues, as well as workflow and Web service technology for execution purposes.

Full text and
other links
PDF (396457 Bytes)
PostScript (2063220 Bytes)
ContactFor more information please contact mariucci@informatik.uni-stuttgart.de
Department(s)University of Stuttgart, Institute of Parallel and Distributed Systems, Applications of Parallel and Distributed Systems
Project(s)ARSENAL
Entry dateSeptember 18, 2003
   Publ. Institute   Publ. Computer Science