Title
Software component composition based on ADL and Middleware.
Abstract
How to compose prefabricated components is a key issue in component-based reuse. Research on Software Architecture (SA) and Component-based Software Development (CBSD) provides two hopeful solutions from different perspectives. SA provides a top-down approach to realizing component-based reuse. However, it pays insufficient attention to the refinement and implementation of the architectural descriptions, and does not provide the necessary capability to automate the transformation or composition to form a final executable application. CBSD provides a bottom-up way by using existing middleware infrastructures. However, these technologies do not take into account the systematic methodology that can guide the CBSD process, especially the component composition at higher abstract levels. We argue that it is a natural solution to combine these two approaches. In this paper, an architecture-based component composition approach is presented. In this way, SA description, using mapping rules and mini-tools to narrow the gap between design and implementation, is used as the blueprint and middleware technology as the runtime scaffold for component composition. Our approach presents an ADL, which supports user-defined connectors and has an extensible framework, to specify software architectures. To map a SA description into implementation, it is necessary to map it first to an OO design model described in UML, then to the final implementation. The architectural description can be mapped into source code or executable code by using some ORB conforming to CORBA standard. Also a toolkit is provided to support this approach efficiently.
Year
DOI
Venue
2001
10.1007/BF02713972
Science in China Series F: Information Sciences
Keywords
DocType
Volume
software component,bottom up,software architecture,source code,component based software development,middleware,top down
Journal
44
Issue
Citations 
PageRank 
2
16
1.93
References 
Authors
10
3
Name
Order
Citations
PageRank
Hong Mei13535219.36
Jichuan Chang254726.22
Fuqing Yang335925.76