Fast computation of spatial selections and joins using graphics hardware

Nagender Bandi, Chengyu Sun, Divyakant Agrawal, Amr El Abbadi

Research output: Contribution to journalArticle

Abstract

Spatial database operations are typically performed in two steps. In the filtering step, indexes and the minimum bounding rectangles (MBRs) of the objects are used to quickly determine a set of candidate objects. In the refinement step, the actual geometries of the objects are retrieved and compared to the query geometry or each other. Because of the complexity of the computational geometry algorithms involved, the CPU cost of the refinement step is usually the dominant cost of the operation for complex geometries such as polygons. Although many run-time and pre-processing-based heuristics have been proposed to alleviate this problem, the CPU cost still remains the bottleneck. In this paper, we propose a novel approach to address this problem using the efficient rendering and searching capabilities of modern graphics hardware. This approach does not require expensive pre-processing of the data or changes to existing storage and index structures, and is applicable to both intersection and distance predicates. We evaluate this approach by comparing the performance with leading software solutions. The results show that by combining hardware and software methods, the overall computational cost can be reduced substantially for both spatial selections and joins. We integrated this hardware/software co-processing technique into a popular database to evaluate its performance in the presence of indexes, pre-processing and other proprietary optimizations. Extensive experimentation with real-world data sets show that the hardware-accelerated technique not only outperforms the run-time software solutions but also performs as well if not better than pre-processing-assisted techniques.

Original languageEnglish
Pages (from-to)1073-1100
Number of pages28
JournalInformation Systems
Volume32
Issue number8
DOIs
Publication statusPublished - 1 Dec 2007
Externally publishedYes

Fingerprint

Hardware
Processing
Program processors
Geometry
Costs
Computational geometry
Join
Software
Data base

Keywords

  • Databases
  • Geographic information systems
  • Query optimization

ASJC Scopus subject areas

  • Management Information Systems
  • Management of Technology and Innovation
  • Hardware and Architecture
  • Information Systems
  • Software

Cite this

Fast computation of spatial selections and joins using graphics hardware. / Bandi, Nagender; Sun, Chengyu; Agrawal, Divyakant; El Abbadi, Amr.

In: Information Systems, Vol. 32, No. 8, 01.12.2007, p. 1073-1100.

Research output: Contribution to journalArticle

Bandi, Nagender ; Sun, Chengyu ; Agrawal, Divyakant ; El Abbadi, Amr. / Fast computation of spatial selections and joins using graphics hardware. In: Information Systems. 2007 ; Vol. 32, No. 8. pp. 1073-1100.
@article{0e28e25934f244fd9fbac238769849c3,
title = "Fast computation of spatial selections and joins using graphics hardware",
abstract = "Spatial database operations are typically performed in two steps. In the filtering step, indexes and the minimum bounding rectangles (MBRs) of the objects are used to quickly determine a set of candidate objects. In the refinement step, the actual geometries of the objects are retrieved and compared to the query geometry or each other. Because of the complexity of the computational geometry algorithms involved, the CPU cost of the refinement step is usually the dominant cost of the operation for complex geometries such as polygons. Although many run-time and pre-processing-based heuristics have been proposed to alleviate this problem, the CPU cost still remains the bottleneck. In this paper, we propose a novel approach to address this problem using the efficient rendering and searching capabilities of modern graphics hardware. This approach does not require expensive pre-processing of the data or changes to existing storage and index structures, and is applicable to both intersection and distance predicates. We evaluate this approach by comparing the performance with leading software solutions. The results show that by combining hardware and software methods, the overall computational cost can be reduced substantially for both spatial selections and joins. We integrated this hardware/software co-processing technique into a popular database to evaluate its performance in the presence of indexes, pre-processing and other proprietary optimizations. Extensive experimentation with real-world data sets show that the hardware-accelerated technique not only outperforms the run-time software solutions but also performs as well if not better than pre-processing-assisted techniques.",
keywords = "Databases, Geographic information systems, Query optimization",
author = "Nagender Bandi and Chengyu Sun and Divyakant Agrawal and {El Abbadi}, Amr",
year = "2007",
month = "12",
day = "1",
doi = "10.1016/j.is.2006.12.001",
language = "English",
volume = "32",
pages = "1073--1100",
journal = "Information Systems",
issn = "0306-4379",
publisher = "Elsevier Limited",
number = "8",

}

TY - JOUR

T1 - Fast computation of spatial selections and joins using graphics hardware

AU - Bandi, Nagender

AU - Sun, Chengyu

AU - Agrawal, Divyakant

AU - El Abbadi, Amr

PY - 2007/12/1

Y1 - 2007/12/1

N2 - Spatial database operations are typically performed in two steps. In the filtering step, indexes and the minimum bounding rectangles (MBRs) of the objects are used to quickly determine a set of candidate objects. In the refinement step, the actual geometries of the objects are retrieved and compared to the query geometry or each other. Because of the complexity of the computational geometry algorithms involved, the CPU cost of the refinement step is usually the dominant cost of the operation for complex geometries such as polygons. Although many run-time and pre-processing-based heuristics have been proposed to alleviate this problem, the CPU cost still remains the bottleneck. In this paper, we propose a novel approach to address this problem using the efficient rendering and searching capabilities of modern graphics hardware. This approach does not require expensive pre-processing of the data or changes to existing storage and index structures, and is applicable to both intersection and distance predicates. We evaluate this approach by comparing the performance with leading software solutions. The results show that by combining hardware and software methods, the overall computational cost can be reduced substantially for both spatial selections and joins. We integrated this hardware/software co-processing technique into a popular database to evaluate its performance in the presence of indexes, pre-processing and other proprietary optimizations. Extensive experimentation with real-world data sets show that the hardware-accelerated technique not only outperforms the run-time software solutions but also performs as well if not better than pre-processing-assisted techniques.

AB - Spatial database operations are typically performed in two steps. In the filtering step, indexes and the minimum bounding rectangles (MBRs) of the objects are used to quickly determine a set of candidate objects. In the refinement step, the actual geometries of the objects are retrieved and compared to the query geometry or each other. Because of the complexity of the computational geometry algorithms involved, the CPU cost of the refinement step is usually the dominant cost of the operation for complex geometries such as polygons. Although many run-time and pre-processing-based heuristics have been proposed to alleviate this problem, the CPU cost still remains the bottleneck. In this paper, we propose a novel approach to address this problem using the efficient rendering and searching capabilities of modern graphics hardware. This approach does not require expensive pre-processing of the data or changes to existing storage and index structures, and is applicable to both intersection and distance predicates. We evaluate this approach by comparing the performance with leading software solutions. The results show that by combining hardware and software methods, the overall computational cost can be reduced substantially for both spatial selections and joins. We integrated this hardware/software co-processing technique into a popular database to evaluate its performance in the presence of indexes, pre-processing and other proprietary optimizations. Extensive experimentation with real-world data sets show that the hardware-accelerated technique not only outperforms the run-time software solutions but also performs as well if not better than pre-processing-assisted techniques.

KW - Databases

KW - Geographic information systems

KW - Query optimization

UR - http://www.scopus.com/inward/record.url?scp=34548494851&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34548494851&partnerID=8YFLogxK

U2 - 10.1016/j.is.2006.12.001

DO - 10.1016/j.is.2006.12.001

M3 - Article

AN - SCOPUS:34548494851

VL - 32

SP - 1073

EP - 1100

JO - Information Systems

JF - Information Systems

SN - 0306-4379

IS - 8

ER -