Abstract
The problem of building large, reliable software systems in a controlled, cost-effective way, the so-called software crisis problem, is one of computer science's great challenges. From the very outset of computing as science, software reuse has been touted as a means to overcome the software crisis issue. Over three decades later, the software community is still grappling with the problem of building large reliable software systems in a controlled, cost effective way; the software crisis problem is alive and well. Today, many computer scientists still regard software reuse as a very powerful vehicle to improve the practice of software engineering. The advantage of amortizing software development cost through reuse continues to be a major objective in the art of building software, even though the tools, methods, languages, and overall understanding of software engineering have changed significantly over the years.
Our work is primarily focused on the development of an Adaptive Application Integration Architecture Framework. Without good integration tools and techniques, reuse is difficult and will probably not happen to any significant degree. In the development of the adaptive integration architecture framework, the primary enabling concept is object-oriented design supported by the unified modeling language. The concepts of software architecture, design patterns, and abstract data views are used in a structured and disciplined manner to established a generic framework. This framework is applied to solve the Enterprise Application Integration (EM) problem in the telecommunications operations support system (OSS) enterprise marketplace.
The proposed adaptive application integration architecture framework facilitates application reusability and flexible business process re-engineering. The architecture addresses the need for modern businesses to continuously redefine themselves to address changing market conditions in an increasingly competitive environment. We have developed a number of Enterprise Application Integration design patterns to enable the implementation of an EAI framework in a definite and repeatable manner. The design patterns allow for integration of commercial off-the-shelf applications into a unified enterprise framework facilitating true application portfolio interoperability. The notion of treating application services as infrastructure services and using business processes to combine them arbitrarily provides a natural way of thinking about adaptable and reusable software systems.
We present a mathematical formalism for the specification of design patterns. This specification constitutes an extension of the basic concepts from many-sorted algebra. In particular, the notion of signature is extended to that of a vector, consisting of a set of linearly independent signatures. The approach can be used to reason about various properties including efforts for component reuse and to facilitate complex largescale software development by providing the developer with design alternatives and support for automatic program verification.
Graduation Date
2001
Semester
Summer
Advisor
Orooji, Ali
Degree
Doctor of Philosophy (Ph.D.)
College
College of Engineering and Computer Science
Department
School of Electrical Engineering and Computer Science
Degree Program
Electrical Engineering and Computer Science
Format
Language
English
Rights
Written permission granted by copyright holder to the University of Central Florida Libraries to digitize and distribute for nonprofit, educational purposes.
Length of Campus-only Access
None
Access Status
Doctoral Dissertation (Open Access)
Identifier
DP0000786
STARS Citation
Williams, Denver Robert Edward, "An Adaptive Integration Architecture for Software Reuse" (2001). Retrospective Theses and Dissertations. 1400.
https://stars.library.ucf.edu/rtd/1400
Contributor (Linked data)
University of Central Florida. College of Engineering and Computer Science (Q7895235)
University of Central Florida. College of Engineering and Computer Science [VIAF]
University of Central Florida. College of Engineering and Computer Science [LC]
Accessibility Status
Searchable text