Browse Prior Art Database

Automatic Data Base Design

IP.com Disclosure Number: IPCOM000120406D
Original Publication Date: 1991-Apr-01
Included in the Prior Art Database: 2005-Apr-02
Document File: 2 page(s) / 80K

Publishing Venue

IBM

Related People

Loe, PA: AUTHOR [+2]

Abstract

A software technique is disclosed that simplifies the user's problems in designing a suitable relational data base to meet the requirements of input and output documents.

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 52% of the total text.

Automatic Data Base Design

      A software technique is disclosed that simplifies the
user's problems in designing a suitable relational data base to meet
the requirements of input and output documents.

      In order to maximize the usability and optimize the performance
of a relational data base, it is imperative that the content of each
table is designed according to a set of standard rules.  These cover
such things as unique keys, unique names, and normalized data (third
normal form at least).   The rules have been documented in numerous
industry-standard publications.  There are two ways that the rules
can be applied.  The first way is to rely on the experience and
expertise of a data base designer or systems analyst.  The other way
is to use one of the many commercially available CASE (Computer Aided
Systems Engineering) tools.  Both approaches have drawbacks.  The
first approach has the drawback that such skills are in short supply
and are expensive.  The second has the drawback that the user has to
supply information about all the relationships between all the data
elements which could be a laborious task.

      This article covers a different approach to capturing the
information about the data relationships and an algorithm for
constructing a database design from the data relationships.  The data
relationships are constructed not by ASKING the user to define them
but, instead, by getting the user to merely TELL the system which
data occurs on all his input and output forms, documents, reports and
charts. He just defines what he wants - he does not then need to
extract from these forms all the data elements and the relationships.
These are discovered from an analysis of the forms, which can be
defined in a user-friendly way.  The process described here can
access the form definitions and design an appropriate database.  The
overall simplified process comprises the ordered steps listed below:
1.   Analyze the column names from the forms...