Generating highly customizable SQL parsers

摘要

This first edition of the Software Engineering for Tailor-made Data Management (SETMDM) workshop was held in Nantes, France, March 29, 2008 in conjunction with 11th International Conference on Extending Database Technology (EDBT 2008). >The goal of this workshop was to bring software engineers and database experts together to discuss about technologies for developing tailor-made data management solutions. The workshop was a full day meeting with four paper sessions. It was introduced by a presentation of the FAME-DBMS project that aims at developing tailor-made data management for embedded systems. Another motivating presentation about a tailor-made database system needed to implement advanced file management functionality concluded the first session. In the second session two papers about tailoring transaction processing and lock protocols were presented. The presented solutions allow a specialization of database technologies for a certain application scenario like XML databases. The first presentation of the next session highlighted the issue of the continuously increasing power consumption in current large computing centers. It concluded that also software systems like DBMS should help to decrease power consumption. Using specialized algorithms to tailor data-management can help to improve the energy efficiency. The next paper focused on the complexity of SQL and presented an approach of decomposing SQL-2003 to obtain any required tailor-made subset of the grammar as a basis for a tailor-made SQL parser. The last two papers concentrated on configurability of parts of database management systems by introducing service-oriented technologies and runtime adaptation. Long discussions about services in databases raised in this session. The workshop concluded with an interesting discussion about the future of database system and the well known statement 'one size fits all'. Pros and cons of this topic were lively discussed.