Detecting redundancy in data warehouse evolution

Research output: Chapter in Book/Report/Conference proceedingConference contribution

9 Scopus citations

Abstract

A Data Warehouse (DW) can be abstractly seen as a set of materialized views defined over a set of remote data sources. A DW is intended to satisfy a set of queries. The views materialized in a DW relate to each other in a complex manner, through common sub expressions, in order to guarantee high query performance and low view maintenance cost. DWs are time varying. As time passes new materialized views are added in order to satisfy new queries or for performance reasons while old queries are dropped. The evolution of a DW can result in a redundant set of materialized views. In this paper we address the problem of detecting redundant views in a given DW view selection, that is, views that can be removed from the DW without negatively affecting the query evaluation or the view maintenance process. Using an AND/OR dag representation for multiple queries and views, we first provide a method for detecting materialized views that are not needed in the process of propagating source relation changes to the DW. Then, we use this method to detect materialized views that are redundant. As a side effect, our approach shows how source relation changes can be propagated to the DW materialized views by exploiting common sub expressions between views and by using other materialized views that are not affected by these changes.

Original languageEnglish (US)
Title of host publicationConceptual Modeling – ER 1999 - 18th International Conference on Conceptual Modeling, Proceedings
EditorsJacky Akoka, Mokrane Bouzeghoub, Elisabeth Metais, Isabelle Comyn-Wattiau
PublisherSpringer Verlag
Pages340-355
Number of pages16
ISBN (Print)3540666869, 9783540666868
DOIs
StatePublished - 1999
Externally publishedYes
Event18th International Conference on Conceptual Modeling, ER 1999 - Paris, France
Duration: Nov 15 1999Nov 18 1999

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume1728
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other18th International Conference on Conceptual Modeling, ER 1999
Country/TerritoryFrance
CityParis
Period11/15/9911/18/99

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • General Computer Science

Fingerprint

Dive into the research topics of 'Detecting redundancy in data warehouse evolution'. Together they form a unique fingerprint.

Cite this