Metric based framework for Testing & Evaluation of Service-Oriented Systems

Abstract

The increase in significance of service orientation in system development is accelerating with increase in demand for qualitative and cost-effective systems. Service-Oriented Architecture (SOA) is one of the established structural designs used for developing and implementing flexible, reusable, rapid and low-cost service-oriented systems. The established testing and evaluation methods dont work well for systems that are made-up of services (service-oriented system). As a result, several testing and evaluation metrics for service-oriented systems were proposed. However, these metrics were created based on preceding software development approaches that offer insufficient focus to service oriented systems thereby lacking the efficiency to evaluate these systems. Furthermore, Lack of access to source code also frustrates classical mutation-testing approaches, which require seeding the code with errors. This paper discusses different testing and evaluation metrics available for SOS and proposed a theory-grounded framework for testing and evaluation of service-oriented systems with the aim of decreasing cost and increasing the quality of the SOS. Then, the proposed framework is validated theoretically to check its usability and applicability for testing and evaluation of SOS. The results show that the proposed framework is able to decreasing cost and increasing the quality of the SOS.