WAVE: Automatic Verification of Data-Driven Web Services.
Alin Deutsch, Victor Vianu · 2008
Data-driven Web services, viewed broadly as interactive systems available on the Web for users and programs, provide the backbone for increasingly complex Web applications. While this yields everincreasing functionality, the added complexity renders such applications more vulnerable to bugs and failures, potentially compromising their robustness and correctness. Therefore, there is a need to develop verification techniques for such Web services. The WAVE project at UC San Diego aims to develop new approaches for automatic verification of data-driven Web services. The work relies on a novel, highly effective marriage of model checking and database techniques. We summarize briefly the main contributions of the project, which range from theoretical foundations to the successful implementation of a prototype verifier. 1 Verification of stand-alone data-driven Web services We first outline our results on verification of data-driven Web services for single peers in isolation, then dicsuss extensions of the results to compositions of Web services. We focus on services interacting with external users or programs through a Web browser interface, and accessing an underlying database. Such services include e-commerce sites, scientific and other domain-specific portals, e-government, etc. These Web sites are often