Index nesting – an efficient approach to indexing in object-oriented databases

  • Authors:

  • Beng Chin Ooi;Jiawei Han;Hongjun Lu;Kian Lee Tan

  • Affiliations:

  • Department of Information Systems and Computer Science, National University of Singapore, Kent Ridge, Singapore 119260/ e-mail: {ooibc,luhj,tankl}@iscs.nus.sg;School of Computing Science, Simon Fraser University, British Columbia, Canada V5A 1S6/ e-mail: han@cs.sfu.ca;Department of Information Systems and Computer Science, National University of Singapore, Kent Ridge, Singapore 119260/ e-mail: {ooibc,luhj,tankl}@iscs.nus.sg;Department of Information Systems and Computer Science, National University of Singapore, Kent Ridge, Singapore 119260/ e-mail: {ooibc,luhj,tankl}@iscs.nus.sg

  • Venue:
  • The VLDB Journal — The International Journal on Very Large Data Bases
  • Year:
  • 1996

Quantified Score

Hi-index 0.00

Visualization

Abstract

In object-oriented database systems where the concept of the superclass-subclass is supported, an instance of a subclass is also an instance of its superclass. Consequently, the access scope of a query against a class in general includes the access scope of all its subclasses, unless specified otherwise. An index to support superclass-subclass relationship efficiently must provide efficient associative retrievals of objects from a single class or from several classes in a class hierarchy. This paper presents an efficient index called the hierarchical tree (the H-tree). For each class, an H-tree is maintained, allowing efficient search on a single class. These H-trees are appropriately linked to capture the superclass-subclass relationships, thus allowing efficient retrievals of instances from a class hierarchy. Both experimental and analytical results indicate that the H-tree is an efficient indexing structure.