An Approach to Data Consistency Checking for the Dynamic Replacement of Service Process

Abstract

The technology of service-oriented architecture (SOA) and Web service provides a promising method to rapidly develop the distributed interoperable system for E-commerce applications at low cost. Due to the cross-organizational feature, the dynamic replacement of service process is an important way to guarantee the correctness of service software evolution, supporting the capacity of handling unexpected service failures. However, considering the new services used in the reconfigured process may be incompatible, data consistency should be checked to ensure that the original business logics can be accurately simulated. In response to this problem, we propose an approach to data consistency checking for the dynamic replacement of service process. First, the behavioral model and data model for formalizing service process are introduced, and then the data consistency problem of dynamic replacement is discussed. Second, the data replacement patterns are presented to specify different kinds of replacement behaviors. Third, the single-service replacement method is used to check the data consistency of an individual service dynamic replacement, which aims to compute the similarity degree between interface data. The multi-services replacement method is employed to check the data consistency of a set of services dynamic replacements, which explores critical paths to substitute a part of service processes. Finally, a case study and experiments demonstrate the effectiveness and efficiency of the proposed methods in the data consistency of dynamic replacement. Our approach provides fundamental theory guidance to enhance the credibility of service process in the modern service industry.