An Automated Service Identification Method
Pooyan Jamshidi, Saeed Mansour, Kavan Sedighiani, Soheil Jamshidi, Fereidoon Shams · 2012
Context: A key activity in the construction of any enterprise-wide service-oriented system is the identification of enterpriselevel services from business models based on required quality attributes. Conventional service-identification methods rely heavily on the experiences of architects to direct them in the identification of services and architectural elements by descriptive qualitative guidelines. These service-identification methods have proved to be laborious and less productive, given the mere large scale of enterprises and the human limitations in comprehending the mostly non-quantitative and textual service requirements of such large enterprises while deriving proper services out of business models. Objective: In this paper, we present a novel approach called ASIM (Automated Service Identification Method) for automatically identifying and partly specifying enterprise-level software services from business models using best practices and principles of model-driven software development. Method: We have formulated service identification as a multi-objective optimization problem and solved it by a novel metaheuristic optimization algorithm that derives appropriate service abstractions by using appropriate quantitative measures for granularity, coupling, cohesion, reusability, and maintainability. Results: We have developed a tool based on this approach and report a case study carried out by using this tool. Conclusion: ASIM helps architects to derive right architectural elements of service-oriented solutions that in turn lead to effective service models. Furthermore, the introduced method automates one of the labor-intensive tasks and it enables architects to focus on more conceptual activities, which need human intelligence, and eventually result in the development of an efficient service-based solutions.