Return to search

A Framework for Records Management in Relational Database Systems

The problem of records retention is often viewed as simply deleting records when they have outlived their purpose. However, in the world of relational databases there is no standardized notion of a business record and its retention obligations. Unlike physical documents such as forms and reports, information in databases is organized such that one item of data may be part of various legal records and consequently subject to several (and possibly conflicting) retention policies. This thesis proposes a framework for records retention in relational database systems. It presents a mechanism through which users can specify a broad range of protective and destructive data retention policies for relational records. Compared to naïve solutions for enforcing records management policies, our framework is not only significantly more efficient but it also addresses several unanswered questions about how policies can be mapped from given legal requirements to actions on relational data. The novelty in our approach is that we defined a record in a relational database as an arbitrary logical view, effectively allowing us to reduce several challenges in enforcing data retention policies to well-studied problems in database theory. We argue that our expression based approach of tracking records management obligations is not only easier for records managers to use but also far more space/time efficient compared to traditional metadata approaches discussed in the literature. The thesis concludes with a thorough examination of the limitations of the proposed framework and suggestion for future research in the area of records management for relational database management systems.

Identiferoai:union.ndltd.org:WATERLOO/oai:uwspace.uwaterloo.ca:10012/3672
Date02 May 2008
CreatorsAtaullah, Ahmed Ayaz
Source SetsUniversity of Waterloo Electronic Theses Repository
LanguageEnglish
Detected LanguageEnglish
TypeThesis or Dissertation

Page generated in 0.0194 seconds