Towards Dynamic Reconfiguration of a Composite Web Service: An Approach Based on QoS Prediction

Service-oriented architecture provides the ability to combine several web services in order to fulfil a user-specific requirement. In dynamic environments, the appearance of several unforeseen events can destabilize the composite web service (CWS) and affect its quality. To deal with these issues, the composite web service must be dynamically reconfigured. Dynamic reconfiguration may be enhanced by avoiding the invocation of degraded web services by predicting QoS for the candidate web service. In this paper, we propose a dynamic reconfiguration method based on HMM (Hidden Markov Model) states to predict the imminent degradation in QoS and prevent the invocation of partner web services with degraded QoS values. PSO (Particle Swarm Optimization) and SFLA (Shuffled Frog Leaping Algorithm) are used to improve the prediction efficiency of HMM. Through extensive experiments on a real-world dataset, WS-Dream, the results demonstrate that the proposed approach can achieve better prediction accuracy. Moreover, we carried out a case study where we revealed that the proposed approach outperforms several state-of-the-art methods in terms of execution time.

Bertrand Oligopoly Competition in Composite Web Service Network

Because of its heterogeneous nature, a web service can be composed of multiple composite web services. To improve profitability in the software-component economy, software as a service (SaaS) service providers compete more at the composite-service level than at the single-service level. Moreover, because of the collaborative environment, composite web service networks determine both the applicability of the web service and its expected economic behavior. Based on the traditional linear demand model, this article presents a congestion-aware demand model that makes several assumptions regarding the SaaS service marketplace. Then, it formulates the SaaS service providers' pricing behaviors as a network Bertrand oligopoly competition. Key game-theoretic analysis includes the existence and uniqueness of the pure strategy Nash equilibrium. Moreover, this article provides one sufficient condition, where if all SaaS service providers follow the best response strategy, the strategy profile converges to the unique pure strategy Nash equilibrium.

Managing QoS Degradation of Component Web Services in a Dynamic Environment

In Services Oriented Computing, a composite web service is a value-added service composed of loosely coupled, independent, and distributed component web services. Component or partner web services jointly contribute to fulfil functional as well as non-functional requirements of users of a composite web service. One of the fundamental challenges in Services Oriented Computing is to ensure that a composite web service is flexible enough to react to changes (QoS degradation) in its partner web services at the time of execution. In this context, it is important that the time to adapt to changes should not be significant. Several solutions exist for run-time monitoring of partner web services so that clients can replace them with better alternatives when their QoS values degrade. But these solutions follow either a reactive approach (which is time consuming), or a prediction-based proactive approach (again time consuming, and moreover predicted events may never happen). This article proposes a solution using a publish/subscribe mechanism which is distributed between web service clients and the service providers, and follows a proactive preventive approach. It uses mobile agents to communicate partner web service's QoS status to its clients just in time, in order to decide to choose an alternative in case the QoS values are not satisfactory. The prototype is implemented using JAVA and Java Agent DEvelopment framework (JADE) programming languages. The experimental results show effectiveness of the proposed approach when compared with a static approach (the benchmark), as well as with a reactive solution. Moreover, the framework performs well even in the wake of the increasing levels of QoS degradation of partner web services.