Indexing multiversion data warehouse: from ROWID-Based multiversion join index to bitmap-based multiversion join index

  • Authors:
  • Jan Chmiel

  • Affiliations:
  • Institute of Computing Science, Poznań University of Technology

  • Venue:
  • ADBIS'09 Proceedings of the 13th East European conference on Advances in Databases and Information Systems
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper presents two index structures, called a Bitmap-based Multiversion Join Index (B-MVJI), designed for the optimization of star queries that access multiple data warehouse versions. The B-MVJI indexes a two-dimensional space data values - data warehouse versions by means of bitmaps. The variant of the B-MVJI, called BS-MVJI, based on sorted bitmaps is also presented. The B-MVJI and BS-MVJI were evaluated experimentally and compared to some alternative approaches.