Relational modelling for ADABAS implementations at the Australian Bureau of Statistics

Article type: Research Article

Authors: Williams, N.J.R.¹

Note: [1] Noel Williams is currently Assistant Director, Applications Software Section of the Australian Bureau of Statistics. He is responsible for the evaluation, design, implementation and support of generalized software packages for sample selection, collection dispatch, collection control, survey estimation, time series management and statistical analysis. He has a long standing interest in the development and use of relational modelling techniques for data base design. Mr. Williams holds a BEc (Hons) from the Australian National University.

Abstract: The Australian Bureau of Statistics (ABS) makes extensive use of relational modelling to do database design work during systems development. The Bureau's approach to data modelling has evolved over the past seven years, and is now an integral part of ABS methodology for conducting data analysis. In the course of this methodology, a relational model for the data of a particular application system is developed, refined and used as a medium for the specification of usage, backup and security requirements. The simplicity and familiarity of relational modelling concepts allows subject matter and other non-database specialists to play a very active role in database design. The methodology has some useful inbuilt checks and balances to help ensure the quality of the resulting data model design. Several ABS applications have found that use of non-relational facilities such as periodic groups and multiple value fields is essential in obtaining feasible ADABAS implementations of their relational model designs. However some quite direct implementations have been attempted successfully. The ABS is utilizing and developing a number of ADABAS implementation techniques which achieve economies in storage space and accessing i/o without unduly comprising the relational design.

DOI: 10.3233/SJU-1984-2207

Journal: Statistical Journal of the United Nations Economic Commission for Europe, vol. 2, no. 2, pp. 191-201, 1984