Evaluating the Design of Service-Based Applications

Keywords

Service-Based Applications (SBAs), Quality of Service (QoS), Quality Model.

Description

As any complex software systems, Service-Based Applications (SBAs) evolve to fit new user requirements (in terms of quality of service) and new execution contexts (installation of new devices, technologies, or protocols). Thus, new services appear (or are deployed), disappear; and are modified (or expired). These successive changes deteriorate the design of services. However, these changes can have a negative impact on the quality of service (QoS) of SBAs.

The main goal of this project is to propose an approach for evaluating and improving the design of services to ensure a QoS that satisfies user requirements and execution contexts.

To achieve this goal, here are the challenges to raise: (1) Identify automatically good and bad design practices of SBAs to evaluate the design of services; (2) Propose automatically refactorings to correct design defects introduced by developers to improve services and their composition; and (3) Evaluate the impact of defects and refactorings on the QoS of SBAs with a quality model.

The approach will be validated by performing experiments on ubiquitous systems such as the project for home automation developed within the team and called EnTiMid. This validation will be based on different quality attributes defined on the quality mode.

Bibliography

E. Thomas, Service-Oriented Architecture : Concepts, Technology, and Design. Upper Saddle River, NJ, USA : Prentice Hall PTR, 2005. 15

G. Nain, E. Daubert, O. Barais, and J.-M. Jézéquel, "Using mde to build a schizofrenic middleware for home/building automation," in Proceedings of ServiceWave'08 : Networked European Software & Services Initiative (NESSI) Conference, Madrid, Spain, Dec. 2008. 15, 16

S-Cube, \Survey of quality related aspects relevant for service-based applications," S-Cube, Tech. Rep. PO-JRA-1.3.1, July 2008. [Online]. Available : http://www.s-cube-network.eu/results/results-collections/wp-jra-1.3-deliverables-qos-and-sla 16

Working Environment

• Laboratories:
• IRISA, EPI Triskell (Campus de Beaulieu, Université de Rennes 1, France)

• Scientific Advisors:
• Naouel Moha moha@irisa.fr
• Benoit Baudry bbaudry@irisa.fr
• Jean-Marc Jézéquel jezequel@irisa.fr