Abstract not found

Constraint validation has bcc?n difficult to imple-ment efficiently. The major reason for this difficulty lies in the state-dependent nature of integrity constraints and the rt~quiremcnt of both high-level spc&fication and cfficirnt runtimc cnforccmcnt. In this paper, we pro-pose a constraint reformulation approach to rfficicnt constraint validation. We also demonstrate how this knowledge-basrd constraint rcfornmlation can be natu-rally accomplished in the gcncral framework of problem reformulation with the technique of antecedent deriva-tion. We formalize thr reformulation of an integrity constraint as a tree-starch process where the search space is thtr set of all semantic-equivalent alternatives of the original constraint. We also develop control strate-gies and mcta-level rules for carrying out the search c?fficicntly. The major contribution of this work is a new promising approach to cfficirnt constraint valida-tiun and a general framework to accomplish it. 1.

One of the important means of specifying the semantics about data is via integrity constraints. Experience has shown that the conventional database approach to integrity constraint enforcement is not successful. In this paper, we demonstrate the feasibility and power of a knowledge-based approach to the efficiency problem of constraint validation. We propose a transformational mechanism which exploits knowledge about the application domain and database organization to reformulate integrity constraints into semantically equivalent ones from which efficient code can be generated.