Authors:
George Papastefanatos
1
;
Panos Vassiliadis
2
;
Alkis Simitsis
3
;
Konstantinos Aggistalis
2
;
Fotini Pechlivani
2
and
Yannis Vassiliou
1
Affiliations:
1
National Technical University of Athens, Greece
;
2
University of Ioannina, Greece
;
3
IBM Almaden Research Center, United States
Keyword(s):
Database schema evolution, SQL extension.
Related
Ontology
Subjects/Areas/Topics:
Coupling and Integrating Heterogeneous Data Sources
;
Data Warehouses and OLAP
;
Databases and Information Systems Integration
;
Enterprise Information Systems
Abstract:
The administrators and designers of modern Information Systems face the problem of maintaining their systems in the presence of frequently occurring changes in any counterpart of it. In other words, when a change occurs in any point of the system –e.g., source, schema, view, software construct– they should propagate the change in all the involved parts of the system. Hence, it is imperative that the whole process should be done correctly, i.e., the change should be propagated to all the appropriate points of the system, with a limited overhead imposed on both the system and the humans, who design and maintain it. In this paper, we are dealing with the problem of evolution in the context of databases. First, we present a coherent, graph-based framework for capturing the effect of potential changes in the database software of an Information System. Next, we describe a generic annotation policy for database evolution and we propose a feasible and powerful extension to the SQL language s
pecifically tailored for the management of evolution. Finally, we demonstrate the efficiency and feasibility of our approach through a case study based on a real-world situation occurred in the Greek public sector.
(More)