|
|
|
@inproceedings{DBLP:conf/ssdbm/DixieW83,
author = {John Dixie and
Philip Wake},
editor = {Roy Hammond and
John L. McCarthy},
title = {A Strategy for Implementing a Computer Efficient Database Management
System - Preliminary Research Report},
booktitle = {Proceedings of the Second International Workshop on Statistical
Database Management, Los Altos, California, USA, September 27-29,
1983},
publisher = {Lawrence Berkeley Laboratory},
year = {1983},
pages = {331-337},
ee = {db/conf/ssdbm/DixieW83.html},
crossref = {DBLP:conf/ssdbm/83},
bibsource = {DBLP, http://dblp.uni-trier.de}
}
BibTeX
The United Kingdom census office has a need to improve access to large data sets such as the population census, registrations of births, marriages, deaths and diseases, and major social surveys. In general terms the need is to access a small number of data fields from a large number of records (with or without filtering) for applications in which the data volume is encormous but stable (say 1OOMb or greater). A DBMS with transposed file structure (like RAPID) would appear to be ideal for this.
A strategy is proposed for implementing such a system on ICL 2900 range computers. Questions are raised concerning enhanceability, programming languages, data packing and file structure, storage of meta-data, and the use of the operating system. A recently implemented secondary (macro) data TDF (Transposed Datastore File) is described and the possibility of using the same structure for primary (micro) data is discussed.