Please use this identifier to cite or link to this item:
Title: Improving Query Efficiency in High Dimensional Point Indexes
Authors: Outsios, Evangelos
Evangelidis, Georgios
Type: Conference Paper
Subjects: FRASCATI::Natural sciences::Computer and information sciences
Keywords: K-dimensional point indexing
Optimizing data node storage utilization
Range query performance
Issue Date: 2012
First Page: 30
Last Page: 33
Volume Title: International Conference on Integrated Information 2011, Island of Kos, Greece
Abstract: In this paper, we focus on the leaf level nodes of tree-like k-dimensional indexes that store the data entries, since those nodes represent the majority of the nodes in the index. We propose a generic node splitting approach that defers splitting when possible and instead favors merging of a full node with an appropriate sibling and then re-splitting of the resulting node. Our experiments with the hB-tree, show that the proposed splitting approach achieves high average node storage utilization regardless of data distribution, data insertion patterns and dimensionality.
Appears in Collections:Department of Applied Informatics

Files in This Item:
File Description SizeFormat 
2011_ICININFO_Outsios.pdf439,24 kBAdobe PDFView/Open

This item is licensed under a Creative Commons License Creative Commons