Improving Service Availability without Improving Availability of Individual Services

This paper presents a novel architectural solution for improving dependability of Web Services. This approach is based on the concepts from the emerging resilienceexplicit computing combined with the traditional fault-tolerance techniques such as recovery blocks and N-version programming applied in the context of the serviceoriented architecture. We propose a distributed solution called WS-Mediator, which is implemented as an overlay network of specialized services. The globally distributed architecture of the WS-Mediator system collects dependability metadata from the end-user’s perspective, analyses them and acts upon them to tolerate faults using dynamic reconfiguration. Therefore to improve dependability of Web Services by introducing service redundancy. We have implemented a Java WS-Mediator framework based upon WS-Mediator concept, which can be easily integrated into implementation of Java Web Services applications. We report the results of the extensive experiments conducted in the context of the bioinformatics domain, in which we demonstrate the applicability of our approach. © 2007 University of Newcastle upon Tyne. Printed and published by the University of Newcastle upon Tyne, Computing Science, Claremont Tower, Claremont Road, Newcastle upon Tyne, NE1 7RU, England. Bibliographical details CHEN, Y., ROMANOVSKY, A. An analysis of the dynamics of British academic science [By] Y. Chen, A. Romanovsky. Newcastle upon Tyne: University of Newcastle upon Tyne: Computing Science, 2007. (University of Newcastle upon Tyne, Computing Science, Technical Report Series, No. CS-TR-1025)