Abstract

The flexibility and dynamism of service-based applications impose shifting the validation process to runtime; therefore, runtime monitoring of dynamic features attached to service-based systems is becoming an important direction of research that motivated the definition of our work. We propose an ontology based contextualization and a framework and techniques for managing context constraints in a Web service process for dynamic requirements validation monitoring at process runtime. Firstly, we propose an approach to define and model dynamic service context attached to composition and execution of services in a service process at run-time. Secondly, managing context constraints are defined in a framework, which has three main processes for context manipulation and reasoning, context constraints generation, and dynamic instrumentation and validation monitoring of context constraints. The dynamic requirements attached to service composition and execution are generated as context constraints. The dynamic service context modeling is investigated based on empirical analysis of application scenarios in the classical business domain and analysing previous models in the literature. The orientation of context aspects in a general context taxonomy is considered important. The Ontology Web Language (OWL) has many merits on formalising dynamic service context such as shared conceptualization, logical language support for composition and reasoning, XML based interoperability, etc. XML-based constraint representation is compatible with Web service technologies. The analysis of complementary case study scenarios and expert opinions through a survey illustrate the validity and completeness of our context model. The proposed techniques for context manipulation, context constraints generation, instrumentation and validation monitoring are investigated through a set of experiments from an empirical evaluation. The analytical evaluation is also used to evaluate algorithms. Our contributions and evaluation results provide a further step towards developing a highly automated dynamic requirements management system for service processes at process run-time.