Statistics For The Usage Of A Conceptual Data Model As A Basis For Logical Data Base Design

• Published in: Fifth International Conference on Very Large Data Bases, 1979 pp. 140 - 145
• Main Authors: Oren, O., Aschim, F.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 1979
• Subjects: Buildings; Data models; Documentation; High level languages; Information systems; Logic programming; Packaging; Statistical analysis; Statistics
• DOI: 10.1109/VLDB.1979.718129

A data base design aid, SYDADA, is introduced. Input to SYDADA is a requirements specification, expressed in a language called SYSDOC. A SYSDOC specification consists of two main parts: a conceptual data model and a transaction description. The conceptual data model is a high level, implementation independent data description. The transaction description is expressed in a very high level language which uses the conceptual data model. SYDADA analyzes the requirements specification by modelling the specified system. In the system model the conceptual data model is the data base, and the transactions specified can be executed using this data base. The most important result of the analysis is a set of statistics showing how, and how frequently, the various parts of the conceptual data model are used. We have shown how these statistics can be used as a basis for logical data base design.