Title :
A Graph Theory Based Impact and Completion Analysis Framework and Applications for Modeling SOA Solution Components
Author :
Zhou, Nianjun ; Zhang, Liang-Jie
Author_Institution :
IBM T.J. Watson Res. Center, Hawthorne, NY
Abstract :
Evolved from our engineering experience, this paper presents a mathematical framework to define and analyze an SOA (service-oriented architecture) model. SOA model, composed by design elements, is represented as directed graph based on graph theory. For each design element, two directed graphs are created to reflect the panoramic view and relationships of this design element with other design elements of same model, and used for impact and completion analysis of this design element. A numerical value called the relative importance indicator is computed to quantify the relationship between any two design elements. This indicator forms a matrix that is used as a base for more advanced analysis, such as model partition, model coupling, and variation-oriented design. Some future research directions such as model reduction are discussed at the end of this paper.
Keywords :
directed graphs; software architecture; completion analysis; directed graph theory; impact analysis; model coupling; model partition; relative importance indicator; service-oriented architecture; variation-oriented design; Abstracts; Algorithm design and analysis; Design engineering; Graph theory; Mathematical model; Pattern analysis; Semiconductor optical amplifiers; Service oriented architecture; USA Councils; Unified modeling language; SOA; graph theory; impact and completion analysis; model partition; modeling; relative importance indicator; variation-oriented design;
Conference_Titel :
Services Computing, 2008. SCC '08. IEEE International Conference on
Print_ISBN :
978-0-7695-3283-7
DOI :
10.1109/SCC.2008.137