|
Abstract : |
The provision of more intelligent support for complex database applications is becoming an important research topic, and declaratively stated integrity constraints are increasingly being seen as a potential source of semantic information for such advanced DBMS's. However, despite much research e#ort, the problem of e#ciently compiling and checking a wide range of complex constraints has not yet been solved. And until it is solved, few database designers will be willing to sacrifice the performance of the every day database usage, in order to provide a complete semantic modelling of their application domain. In this paper, we describe the implementation of a constraint checking architecture and compiler, which have been designed to combine an expressive constraint language with e#cient run-time maintenance. While the current implementation does not yet achieve high e#ciency for all constraints, it has been designed with extensibility in mind and forms a solid platform from which to investigate further techniques for improving the handling of a wider range of constraints. The system is currently being used to describe and enforce the semantics of a complex scientific application domain, namely three-dimensional protein structure., |
