QoS-Driven Self-Healing Web Service Composition Based on Performance Prediction

Abstract

Web services run in a highly dynamic environment, as a result, the QoS of which will change relatively frequently. In order to make the composite service adapt to such dynamic property of Web services, we propose a self-healing approach for web service composition. Such an approach is an integration of backing up in selection and reselecting in execution. In order to make the composite service heal itself as quickly as possible and minimize the number of reselections, a way of performance prediction is proposed in this paper. On this basis, the self-healing approach is presented including framework, the triggering algorithm of the reselection and the reliability model of the service. Experiments show that the proposed solutions have better performance in supporting the self-healing Web service composition.

Web Service Composition in Dynamic Environment : A Comparative Study

“…Moreover, the work does not consider the changes in QoS of participant services. [10] the authors in this research present a self-healing service composition framework based on performance prediction. The framework contains a QoS monitor that collects the quality values for participating services.…”

Fuzzy Logic Based QoS Optimization Mechanism for Service Composition

“…Both parameters have been used in other approaches, like those presented in [3,10,22,24]. The amount of energy used by data centres has not only economical but also environmental impacts.…”

Quality-of-Service-Aware Service Selection: A Novel Approach Considering Potential Service Failures and Nondeterministic Service Values

“…This may also be a reason why the travel domain is widely used to illustrate the problem of QoS-aware service selection (cf. Cardellini et al 2007;Chifu et al 2010;Dai et al 2009;Gao et al 2006;Grossmann et al 2011;Guo et al 2007;Hwang et al 2008;Li et al 2010;Li et al 2011;Mei et al 2008;Yang et al 2009;Zeng et al 2008). Second, a documentation of the service candidates and their corresponding QoS values for the QoS attributes response time and price is provided by Canfora et al (2008).…”