WSEAS Transactions on Computers
Print ISSN: 1109-2750, E-ISSN: 2224-2872
Volume 13, 2014
The Design of Multidimensional Data Model Using Principles of the Anchor Data Modeling: An Assessment of Experimental Approach Based on Query Execution Performance
Authors: ,
Abstract: The decision making processes need to reflect changes in the business world in a multidimensional way. This includes also similar way of viewing the data for carrying out key decisions that ensure competitiveness of the business. In this paper we focus on the Business Intelligence system as a main toolset that helps in carrying out complex decisions and which requires multidimensional view of data for this purpose. We propose a novel experimental approach to the design a multidimensional data model that uses principles of the anchor modeling technique. The proposed approach is expected to bring several benefits like better query execution performance, better support for temporal querying and several others. We provide assessment of this approach mainly from the query execution performance perspective in this paper. The emphasis is placed on the assessment of this technique as a potential innovative approach for the field of the data warehousing with some implicit principles that could make the process of the design, implementation and maintenance of the data warehouse more effective. The query performance testing was performed in the row-oriented database environment using a sample of 10 star queries executed in the environment of 10 sample multidimensional data models. The results show comparison of differences between results of query execution in the environment of the experimental “Anchor” schema and the traditional “Star” schema using statistical methods. The results show possible indications towards expected benefits of the proposed approach that embraces high level of normalization of the resulting database schema in contrast with the traditional approach that results mostly to the creation of a non-normalized database schema.
Search Articles
Keywords: Μultidimensional view of data, multidimensional data model, experimental approach, Anchor modeling, query execution performance
Pages: 177-194
WSEAS Transactions on Computers, ISSN / E-ISSN: 1109-2750 / 2224-2872, Volume 13, 2014, Art. #15