Français Anglais
Accueil Annuaire Plan du site
Research results
Contract / grant for group Formal Testing and System Exploration


Dec 2007 - May 2010

Group : Formal Testing and System Exploration
Principal investigator : ZAIDI Fatiha

Administrator : 
Affiliation : Université Paris XI

Web Service Modelling and Validation

The main objective of WebMov is to contribute to the design, composition and validation of Web Services through a high level of abstraction view and a SOA based logical architecture vision. In this domain, industry usually constructs new services by composition of modules which describe existing Web Services. These composition mechanisms are called orchestration. In this proposal, we are interested in the design and composition mechanisms for Web Services as well as their validation using different types of testing techniques.
WebMov will define logical models for SOA architectures in order to provide a high level of abstraction to design the architecture, the service orchestration and the test specifications and purposes.
To achieve composition, IBM and other companies have defined an executable standarized language, BPEL, that allows the description of Web Services and their composition. We propose to use this language as specification service language. Once the modules are described using BPEL, a mapping to a formal model, close to that of extended finite state machines with temporal constraints will be followed. This formalism will facilitate timed test generation. These tests will be executed following two approaches: active testing, consists of executing the produced tests on the system under test; and, passive testing, consist of installing a probe in the system architecture.
Main web domain stakeholders will profit from the approaches developed in WebMov since it will contribute to the reliability of Web Services. The fact that it is based on widely used standardized languages makes these approaches easier to adopt. It must also be noted that Web Services, when executed, use public resources. Errors or deadlocks in their behaviour have a negative impact on the user community and, thus, eliminating them will benefit not only industry but the community as a whole.

Research activities
  Software Testing

ZAIDI Fatiha

Publications :
More information :
Contracts & grants