Optimizing join index based join processing: a graph partitioning approach

Sivakumar Ravada, Shashi Shekhar, Chang tien Lu, Sanjay Chawla

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

4 Citations (Scopus)

Abstract

The cost of join computation, which uses a join-index in a sequential system with limited buffer space, depends primarily on the page access sequence used to fetch the pages of the base relations. In this paper, we introduce a graph-partitioning model that will minimize the length of the page access sequence thus minimizes the redundant I/O, given a fixed buffer. Experiments with Sequoia 2000 data sets show that, the graph-partitioning method outperforms the existing methods based on sorting and online clustering, particularly for a small number of buffers and high join selectivity.

Original languageEnglish
Title of host publicationProceedings of the IEEE Symposium on Reliable Distributed Systems
PublisherIEEE Comp Soc
Pages302-308
Number of pages7
Publication statusPublished - 1998
Externally publishedYes
EventProceedings of the 1998 IEEE 17th Symposium on Reliable Distributed Systems, SRDS - West Lafayette, IN, USA
Duration: 20 Oct 199823 Oct 1998

Other

OtherProceedings of the 1998 IEEE 17th Symposium on Reliable Distributed Systems, SRDS
CityWest Lafayette, IN, USA
Period20/10/9823/10/98

Fingerprint

Sorting
Processing
Costs
Experiments

ASJC Scopus subject areas

  • Hardware and Architecture

Cite this

Ravada, S., Shekhar, S., Lu, C. T., & Chawla, S. (1998). Optimizing join index based join processing: a graph partitioning approach. In Proceedings of the IEEE Symposium on Reliable Distributed Systems (pp. 302-308). IEEE Comp Soc.

Optimizing join index based join processing : a graph partitioning approach. / Ravada, Sivakumar; Shekhar, Shashi; Lu, Chang tien; Chawla, Sanjay.

Proceedings of the IEEE Symposium on Reliable Distributed Systems. IEEE Comp Soc, 1998. p. 302-308.

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

Ravada, S, Shekhar, S, Lu, CT & Chawla, S 1998, Optimizing join index based join processing: a graph partitioning approach. in Proceedings of the IEEE Symposium on Reliable Distributed Systems. IEEE Comp Soc, pp. 302-308, Proceedings of the 1998 IEEE 17th Symposium on Reliable Distributed Systems, SRDS, West Lafayette, IN, USA, 20/10/98.
Ravada S, Shekhar S, Lu CT, Chawla S. Optimizing join index based join processing: a graph partitioning approach. In Proceedings of the IEEE Symposium on Reliable Distributed Systems. IEEE Comp Soc. 1998. p. 302-308
Ravada, Sivakumar ; Shekhar, Shashi ; Lu, Chang tien ; Chawla, Sanjay. / Optimizing join index based join processing : a graph partitioning approach. Proceedings of the IEEE Symposium on Reliable Distributed Systems. IEEE Comp Soc, 1998. pp. 302-308
@inproceedings{0621fc1bdeab4d3ab751f1795a19508d,
title = "Optimizing join index based join processing: a graph partitioning approach",
abstract = "The cost of join computation, which uses a join-index in a sequential system with limited buffer space, depends primarily on the page access sequence used to fetch the pages of the base relations. In this paper, we introduce a graph-partitioning model that will minimize the length of the page access sequence thus minimizes the redundant I/O, given a fixed buffer. Experiments with Sequoia 2000 data sets show that, the graph-partitioning method outperforms the existing methods based on sorting and online clustering, particularly for a small number of buffers and high join selectivity.",
author = "Sivakumar Ravada and Shashi Shekhar and Lu, {Chang tien} and Sanjay Chawla",
year = "1998",
language = "English",
pages = "302--308",
booktitle = "Proceedings of the IEEE Symposium on Reliable Distributed Systems",
publisher = "IEEE Comp Soc",

}

TY - GEN

T1 - Optimizing join index based join processing

T2 - a graph partitioning approach

AU - Ravada, Sivakumar

AU - Shekhar, Shashi

AU - Lu, Chang tien

AU - Chawla, Sanjay

PY - 1998

Y1 - 1998

N2 - The cost of join computation, which uses a join-index in a sequential system with limited buffer space, depends primarily on the page access sequence used to fetch the pages of the base relations. In this paper, we introduce a graph-partitioning model that will minimize the length of the page access sequence thus minimizes the redundant I/O, given a fixed buffer. Experiments with Sequoia 2000 data sets show that, the graph-partitioning method outperforms the existing methods based on sorting and online clustering, particularly for a small number of buffers and high join selectivity.

AB - The cost of join computation, which uses a join-index in a sequential system with limited buffer space, depends primarily on the page access sequence used to fetch the pages of the base relations. In this paper, we introduce a graph-partitioning model that will minimize the length of the page access sequence thus minimizes the redundant I/O, given a fixed buffer. Experiments with Sequoia 2000 data sets show that, the graph-partitioning method outperforms the existing methods based on sorting and online clustering, particularly for a small number of buffers and high join selectivity.

UR - http://www.scopus.com/inward/record.url?scp=0032313819&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0032313819&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0032313819

SP - 302

EP - 308

BT - Proceedings of the IEEE Symposium on Reliable Distributed Systems

PB - IEEE Comp Soc

ER -