In this paper, we describe our experiences in building and evolving an XQuery engine with a focus on data and service federation use cases. The engine that we discuss is a core component of the BEA AquaLogic Data Services Platform product (recently re-released under the name Oracle Data Service Integrator). This XQuery engine was designed to provide efficient query and update capabilities over various classes of enterprise data sources, serving as the data access layer in a service-oriented architecture (SOA). The goal of this paper is to give an architectural overview of the engine, discussing some of the key implementation techniques that were employed as well as several XQuery language extensions that were introduced to address common data and service integration problems and challenges.
Michael Blow, Vinayak R. Borkar, Michael J. Carey,