A Genetic-Based Adaptive Approach for Reliable and Efficient Service Composition

Composite services often run in dynamically changing environments where new composition requirements can be added and Web services may disappear or their quality of service may decline. The service composition requirements are classified into three groups: functional, nonfunctional, and transactional requirements. The functional computing requirements concern functional capabilities of Web services. The nonfunctional computing requirements are defined as service-level agreements (SLA) constraints. The transactional computing requirements define behavioral constraints and are expressed by designers using the acceptable termination states (ATS) concept. In this work, we proposed an adaptive service composition (ASC) approach. Such an approach is adaptive as it can be used to either define or reconfigure composite services according to SLA and ATS constraints. It introduces a new recovery capability (RC) concept for capturing the recoverability level supported by Web services. We adopted a genetic-based algorithm to model the ASC approach. The choice of such an algorithm is justified by the nature of the problem of our ASC process which belongs to the NP-hard class. We experimentally showed the effectiveness of our ASC approach and empirically studied the service composition RC in particular.