A new parallel CBF method for shared-nothing cluster-based architecture

Jae Woo Chang; Yong Ki Kim

A new parallel CBF method for shared-nothing cluster-based architecture

Jae Woo Chang^*
, Yong Ki Kim

^*Corresponding author for this work

Jeonbuk National University

Research output: Contribution to conference › Conference paper › peer-review

Abstract

To efficiently retrieve high-dimensional data in multimedia database and data warehousing applications, the VA-file and the CBF (Cell-based filtering) method have been proposed. But they show a linear decreasing on performance as the dimensionality. To cope with the problem, we, in this paper, propose a new parallel CBF method for shared-nothing (SN) cluster-based parallel architecture. In addition, we devise data insertion, range query processing, and k-NN query processing algorithms that are suitable for the SN parallel architecture. Finally, we show that our parallel CBF method achieves good retrieval performance in proportion to the number of servers in the SN parallel architecture, compared with the conventional CBF method.

Original language	English
Title of host publication	Proceedings of the Fifteenth IASTED International Conference on Parallel and Distributed Computing and Sytems
Editors	T. Gonzalez
Pages	105-110
Number of pages	6
Edition	1
State	Published - 2003
Event	Proceedings of the Fifteenth IASTED International Conference on Parallel and Distributed Computing and Systems - Marina del Rey, CA, United States Duration: 2003.11.3 → 2003.11.5

Publication series

Name	Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems
Number	1
Volume	15

Conference

Conference	Proceedings of the Fifteenth IASTED International Conference on Parallel and Distributed Computing and Systems
Country/Territory	United States
City	Marina del Rey, CA
Period	03.11.3 → 03.11.5

Keywords

Cluster-based parallel architecture
Parallel CBF method
Shared-nothing architecture

Quacquarelli Symonds(QS) Subject Topics

Computer Science & Information Systems

Cite this

Chang, J. W., & Kim, Y. K. (2003). A new parallel CBF method for shared-nothing cluster-based architecture. In T. Gonzalez (Ed.), Proceedings of the Fifteenth IASTED International Conference on Parallel and Distributed Computing and Sytems (1 ed., pp. 105-110). (Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems; Vol. 15, No. 1).

Chang, Jae Woo ; Kim, Yong Ki. / A new parallel CBF method for shared-nothing cluster-based architecture. Proceedings of the Fifteenth IASTED International Conference on Parallel and Distributed Computing and Sytems. editor / T. Gonzalez. 1. ed. 2003. pp. 105-110 (Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems; 1).

@inproceedings{011da91be49e4575acd7a31e5d339327,

title = "A new parallel CBF method for shared-nothing cluster-based architecture",

abstract = "To efficiently retrieve high-dimensional data in multimedia database and data warehousing applications, the VA-file and the CBF (Cell-based filtering) method have been proposed. But they show a linear decreasing on performance as the dimensionality. To cope with the problem, we, in this paper, propose a new parallel CBF method for shared-nothing (SN) cluster-based parallel architecture. In addition, we devise data insertion, range query processing, and k-NN query processing algorithms that are suitable for the SN parallel architecture. Finally, we show that our parallel CBF method achieves good retrieval performance in proportion to the number of servers in the SN parallel architecture, compared with the conventional CBF method.",

keywords = "Cluster-based parallel architecture, Parallel CBF method, Shared-nothing architecture",

author = "Chang, \{Jae Woo\} and Kim, \{Yong Ki\}",

year = "2003",

language = "English",

isbn = "088986392X",

series = "Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems",

number = "1",

pages = "105--110",

editor = "T. Gonzalez",

booktitle = "Proceedings of the Fifteenth IASTED International Conference on Parallel and Distributed Computing and Sytems",

edition = "1",

note = "Proceedings of the Fifteenth IASTED International Conference on Parallel and Distributed Computing and Systems ; Conference date: 03-11-2003 Through 05-11-2003",

}

Chang, JW & Kim, YK 2003, A new parallel CBF method for shared-nothing cluster-based architecture. in T Gonzalez (ed.), Proceedings of the Fifteenth IASTED International Conference on Parallel and Distributed Computing and Sytems. 1 edn, Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems, no. 1, vol. 15, pp. 105-110, Proceedings of the Fifteenth IASTED International Conference on Parallel and Distributed Computing and Systems, Marina del Rey, CA, United States, 03.11.3.

A new parallel CBF method for shared-nothing cluster-based architecture. / Chang, Jae Woo; Kim, Yong Ki.
Proceedings of the Fifteenth IASTED International Conference on Parallel and Distributed Computing and Sytems. ed. / T. Gonzalez. 1. ed. 2003. p. 105-110 (Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems; Vol. 15, No. 1).

Research output: Contribution to conference › Conference paper › peer-review

TY - GEN

T1 - A new parallel CBF method for shared-nothing cluster-based architecture

AU - Chang, Jae Woo

AU - Kim, Yong Ki

PY - 2003

Y1 - 2003

N2 - To efficiently retrieve high-dimensional data in multimedia database and data warehousing applications, the VA-file and the CBF (Cell-based filtering) method have been proposed. But they show a linear decreasing on performance as the dimensionality. To cope with the problem, we, in this paper, propose a new parallel CBF method for shared-nothing (SN) cluster-based parallel architecture. In addition, we devise data insertion, range query processing, and k-NN query processing algorithms that are suitable for the SN parallel architecture. Finally, we show that our parallel CBF method achieves good retrieval performance in proportion to the number of servers in the SN parallel architecture, compared with the conventional CBF method.

AB - To efficiently retrieve high-dimensional data in multimedia database and data warehousing applications, the VA-file and the CBF (Cell-based filtering) method have been proposed. But they show a linear decreasing on performance as the dimensionality. To cope with the problem, we, in this paper, propose a new parallel CBF method for shared-nothing (SN) cluster-based parallel architecture. In addition, we devise data insertion, range query processing, and k-NN query processing algorithms that are suitable for the SN parallel architecture. Finally, we show that our parallel CBF method achieves good retrieval performance in proportion to the number of servers in the SN parallel architecture, compared with the conventional CBF method.

KW - Cluster-based parallel architecture

KW - Parallel CBF method

KW - Shared-nothing architecture

UR - https://www.scopus.com/pages/publications/1542537895

M3 - Conference paper

AN - SCOPUS:1542537895

SN - 088986392X

T3 - Proceedings of the IASTED International Conference on Parallel and Distributed Computing and Systems

SP - 105

EP - 110

BT - Proceedings of the Fifteenth IASTED International Conference on Parallel and Distributed Computing and Sytems

A2 - Gonzalez, T.

T2 - Proceedings of the Fifteenth IASTED International Conference on Parallel and Distributed Computing and Systems

Y2 - 3 November 2003 through 5 November 2003

ER -

A new parallel CBF method for shared-nothing cluster-based architecture

Abstract

Publication series

Conference

Keywords

Quacquarelli Symonds(QS) Subject Topics

Other files and links

Fingerprint

Cite this