Spatial indexing of large multidimensional databases

I. Csabai, M. Trencseni, G. Herczegh, L. Dobos, P. Jozsa, N. Purger, T. Budavari, A. Szalay

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

4 Citations (Scopus)

Abstract

Scientific endeavors such as large astronomical surveys generate databases on the terabyte scale. These, usually multidimensional databases must be visualized and mined in order to find interesting objects or to extract meaningful and qualitatively new relationships. Many statistical algorithms required for these tasks run reasonably fast when operating on small sets of in-memory data, but take noticeable performance hits when operating on large databases that do not fit into memory. We utilize new software technologies to develop and evaluate fast multidimensional indexing schemes that inherently follow the underlying, highly non-uniform distribution of the data: they are layered uniform grid indices, hierarchical binary space partitioning, and sampled flat Voronoi tessellation of the data. Our working database is the 5-dimensional magnitude space of the Sloan Digital Sky Survey with more than 270 million data points, where we show that these techniques can dramatically speed up data mining operations such as finding similar objects by example, classifying objects or comparing extensive simulation sets with observations. We are also developing tools to interact with the multidimensional database and visualize the data at multiple resolutions in an adaptive manner.

Original languageEnglish
Title of host publicationCIDR 2007 - 3rd Biennial Conference on Innovative Data Systems Research
Pages207-218
Number of pages12
Publication statusPublished - 2007
Event3rd Biennial Conference on Innovative Data Systems Research, CIDR 2007 - Asilomar, CA, United States
Duration: Jan 7 2007Jan 10 2007

Other

Other3rd Biennial Conference on Innovative Data Systems Research, CIDR 2007
CountryUnited States
CityAsilomar, CA
Period1/7/071/10/07

Fingerprint

Data storage equipment
Data mining

Keywords

  • Data mining
  • Kd-tree. voronoi
  • Large databases
  • Multidimensional spatial databases
  • Spatial indexing
  • Visualization

ASJC Scopus subject areas

  • Information Systems

Cite this

Csabai, I., Trencseni, M., Herczegh, G., Dobos, L., Jozsa, P., Purger, N., ... Szalay, A. (2007). Spatial indexing of large multidimensional databases. In CIDR 2007 - 3rd Biennial Conference on Innovative Data Systems Research (pp. 207-218)

Spatial indexing of large multidimensional databases. / Csabai, I.; Trencseni, M.; Herczegh, G.; Dobos, L.; Jozsa, P.; Purger, N.; Budavari, T.; Szalay, A.

CIDR 2007 - 3rd Biennial Conference on Innovative Data Systems Research. 2007. p. 207-218.

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

Csabai, I, Trencseni, M, Herczegh, G, Dobos, L, Jozsa, P, Purger, N, Budavari, T & Szalay, A 2007, Spatial indexing of large multidimensional databases. in CIDR 2007 - 3rd Biennial Conference on Innovative Data Systems Research. pp. 207-218, 3rd Biennial Conference on Innovative Data Systems Research, CIDR 2007, Asilomar, CA, United States, 1/7/07.
Csabai I, Trencseni M, Herczegh G, Dobos L, Jozsa P, Purger N et al. Spatial indexing of large multidimensional databases. In CIDR 2007 - 3rd Biennial Conference on Innovative Data Systems Research. 2007. p. 207-218
Csabai, I. ; Trencseni, M. ; Herczegh, G. ; Dobos, L. ; Jozsa, P. ; Purger, N. ; Budavari, T. ; Szalay, A. / Spatial indexing of large multidimensional databases. CIDR 2007 - 3rd Biennial Conference on Innovative Data Systems Research. 2007. pp. 207-218
@inproceedings{17d2d39a1a594ac7ade13cbec8de3536,
title = "Spatial indexing of large multidimensional databases",
abstract = "Scientific endeavors such as large astronomical surveys generate databases on the terabyte scale. These, usually multidimensional databases must be visualized and mined in order to find interesting objects or to extract meaningful and qualitatively new relationships. Many statistical algorithms required for these tasks run reasonably fast when operating on small sets of in-memory data, but take noticeable performance hits when operating on large databases that do not fit into memory. We utilize new software technologies to develop and evaluate fast multidimensional indexing schemes that inherently follow the underlying, highly non-uniform distribution of the data: they are layered uniform grid indices, hierarchical binary space partitioning, and sampled flat Voronoi tessellation of the data. Our working database is the 5-dimensional magnitude space of the Sloan Digital Sky Survey with more than 270 million data points, where we show that these techniques can dramatically speed up data mining operations such as finding similar objects by example, classifying objects or comparing extensive simulation sets with observations. We are also developing tools to interact with the multidimensional database and visualize the data at multiple resolutions in an adaptive manner.",
keywords = "Data mining, Kd-tree. voronoi, Large databases, Multidimensional spatial databases, Spatial indexing, Visualization",
author = "I. Csabai and M. Trencseni and G. Herczegh and L. Dobos and P. Jozsa and N. Purger and T. Budavari and A. Szalay",
year = "2007",
language = "English",
pages = "207--218",
booktitle = "CIDR 2007 - 3rd Biennial Conference on Innovative Data Systems Research",

}

TY - GEN

T1 - Spatial indexing of large multidimensional databases

AU - Csabai, I.

AU - Trencseni, M.

AU - Herczegh, G.

AU - Dobos, L.

AU - Jozsa, P.

AU - Purger, N.

AU - Budavari, T.

AU - Szalay, A.

PY - 2007

Y1 - 2007

N2 - Scientific endeavors such as large astronomical surveys generate databases on the terabyte scale. These, usually multidimensional databases must be visualized and mined in order to find interesting objects or to extract meaningful and qualitatively new relationships. Many statistical algorithms required for these tasks run reasonably fast when operating on small sets of in-memory data, but take noticeable performance hits when operating on large databases that do not fit into memory. We utilize new software technologies to develop and evaluate fast multidimensional indexing schemes that inherently follow the underlying, highly non-uniform distribution of the data: they are layered uniform grid indices, hierarchical binary space partitioning, and sampled flat Voronoi tessellation of the data. Our working database is the 5-dimensional magnitude space of the Sloan Digital Sky Survey with more than 270 million data points, where we show that these techniques can dramatically speed up data mining operations such as finding similar objects by example, classifying objects or comparing extensive simulation sets with observations. We are also developing tools to interact with the multidimensional database and visualize the data at multiple resolutions in an adaptive manner.

AB - Scientific endeavors such as large astronomical surveys generate databases on the terabyte scale. These, usually multidimensional databases must be visualized and mined in order to find interesting objects or to extract meaningful and qualitatively new relationships. Many statistical algorithms required for these tasks run reasonably fast when operating on small sets of in-memory data, but take noticeable performance hits when operating on large databases that do not fit into memory. We utilize new software technologies to develop and evaluate fast multidimensional indexing schemes that inherently follow the underlying, highly non-uniform distribution of the data: they are layered uniform grid indices, hierarchical binary space partitioning, and sampled flat Voronoi tessellation of the data. Our working database is the 5-dimensional magnitude space of the Sloan Digital Sky Survey with more than 270 million data points, where we show that these techniques can dramatically speed up data mining operations such as finding similar objects by example, classifying objects or comparing extensive simulation sets with observations. We are also developing tools to interact with the multidimensional database and visualize the data at multiple resolutions in an adaptive manner.

KW - Data mining

KW - Kd-tree. voronoi

KW - Large databases

KW - Multidimensional spatial databases

KW - Spatial indexing

KW - Visualization

UR - http://www.scopus.com/inward/record.url?scp=67649653673&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67649653673&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:67649653673

SP - 207

EP - 218

BT - CIDR 2007 - 3rd Biennial Conference on Innovative Data Systems Research

ER -