Improving the Deductive System DES with Persistence by Using SQL DBMS's
This work addresses the need for persistence in deductive systems for users in database and AI domains, but it is incremental as it builds on existing DES and SQL technologies.
The authors tackled the problem of adding persistent predicates to the in-memory deductive system DES by integrating it with external SQL database management systems, enabling deductive expressive power over relational databases and intermixing computations, with a performance analysis showing it compares to current relational database systems.
This work presents how persistent predicates have been included in the in-memory deductive system DES by relying on external SQL database management systems. We introduce how persistence is supported from a user-point of view and the possible applications the system opens up, as the deductive expressive power is projected to relational databases. Also, we describe how it is possible to intermix computations of the deductive engine and the external database, explaining its implementation and some optimizations. Finally, a performance analysis is undertaken, comparing the system with current relational database systems.