Browse Prior Art Database

ADMINS: Data Management On A Minicomputer

IP.com Disclosure Number: IPCOM000128144D
Original Publication Date: 1973-Jun-01
Included in the Prior Art Database: 2005-Sep-15
Document File: 11 page(s) / 44K

Publishing Venue

Software Patent Institute

Related People

Mc Intosh, Stuart: AUTHOR [+4]

Abstract

A powerful and easy-to-use data management system has been implemented on a minicomputer for the managing of multi-million byte complex related data bases stored on disks paks. Managing many files with complex data descriptions and tens of thousands of item lines of actual data occupying tens of millions of bytes of disk storage is completely realistic on a minicomputer, We have developed the logical specifications that could be implemented on any medium or large scale computer, on some minicomputers, and is currently operational on the SEL 7200 minicomputer. Although a secondary objective in our research, we nevertheless consider the development of an applications building environment-- extended algol-like systems language, Cobol, file handling including index sequential and sort, many utilities, flexible execution-- on inexpensive hardware both a significant and highly practical cool in the state of the art of developing inexpensive systems.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 9% of the total text.

Page 1 of 11

THIS DOCUMENT IS AN APPROXIMATE REPRESENTATION OF THE ORIGINAL.

ADMINS: Data Management On A Minicomputer

By Stuart McIntosh and David Griffel

Revised: June, 1973
Center For International Studies
Cambridge, Massachusetts A powerful and easy-to-use data management system has been implemented on a minicomputer 1 for the managing of multi-million byte complex related data bases stored on disks paks. Managing many files with complex data descriptions and tens of thousands of item lines of actual data occupying tens of millions of bytes of disk storage is completely realistic on a minicomputer, We have developed the logical specifications that could be implemented on any medium or large scale computer, on some minicomputers, and is currently operational on the SEL 7200 minicomputer. Although a secondary objective in our research, we nevertheless consider the development of an applications building environment-- extended algol-like systems language, Cobol, file handling including index sequential and sort, many utilities, flexible execution-- on inexpensive hardware both a significant and highly practical cool in the state of the art of developing inexpensive systems.

Data management in ADMINS employs concepts found in large computer data base software as well as Many original concepts. However ADMINS Is particularly suited to a minicomputer environment because ADMINS is very good at partitioning seemin 1y "large" data files according to the logic of hierarchical coding schemes (e.g. the Standard Industrial Classification, e.g. commodity classifications) and tile logic of the use of the data. ADMINS is designed to remove redundancy-- of descriptors (for the same code), of instructions (for the same operation) and of storage (for the same data code). Descriptor lists exist once, even thougi1 they describe characteristics in several files. Instructions exist once, even though they operate on data at different levels (e.g. object data, system log). The storage structures assimilate all the structure present in hierarchical coding schemes and optimize the actual storage of data, removing storage duplications as they occur.

The main ADMINS concepts are:

(1) A Disk based storage structure developed to replace the traditional "logical record" concept used by Cobol and "ISAM" type packages for disk storage of records. Using the special1y developed storage structure and access routines ADMINS is efficiently able to store multiple as well as one value per item, long binary strings and short binary strings as well as single word integers, repetitive patterns of values, text descriptors as well as numerical codes, hierarchical as well as single level codes, relations as well as attributes, and data fragments with missing items and missing categories as well as complete data sets. (2) A data description language for the definition of the relational data structures of many files of related Information such as descriptor records, including relational...