A Physical Design Strategy for Datasets with Multiple Dimensions

View Sample PDF

Author(s): Marlene Goncalves (Universidad Simón Bolívar, Venezuela)and José N. Mendoza (Universidad Simón Bolívar, Venezuela)
Copyright: 2017
Pages: 27
Source title: Intelligent Multidimensional Data Clustering and Analysis
Source Author(s)/Editor(s): Siddhartha Bhattacharyya (RCC Institute of Information Technology, India), Sourav De (Cooch Behar Government Engineering College, India), Indrajit Pan (RCC Institute of Information Technology, India)and Paramartha Dutta (Visva-Bharati University, India)
DOI: 10.4018/978-1-5225-1776-4.ch001

Keywords: Data Analysis & Statistics / Data Analysis and Statistics / Information Science Reference / Library & Information Science

Purchase

View A Physical Design Strategy for Datasets with Multiple Dimensions on the publisher's website for pricing and purchasing information.

Abstract

Since the huge volume of data that is generated today, businesses need to have tools to efficiently manage such data. MonetDB is a column-oriented database management system which has shown to have better query processing time with respect to row-oriented systems. The main objective of this chapter is to propose a physical design strategy that improves query execution times in MonetDB. The proposed physical design strategy was empirically studied for 18 TPC-H queries. The experiments were conducted on the basis of cold cache. Each of the queries were executed first using the proposed physical design strategy in this work and then without any physical design. The reported results show that the runtimes using physical design strategy are better for all queries with a minimum percentage improvement of 29%. Also, they showed that the improvement was statistically significant by means of statistical tests.

The IRMA Community

Research IRM

A Physical Design Strategy for Datasets with Multiple Dimensions

Purchase

Abstract

Related Content

IRMA Sponsors