Abstract
The paper deals with the assessment of an experimental data modeling approach which is intended to support the agile oriented data modeling. The approach is based on the Anchor Data Modeling technique and is applied on a multidimensional data model. The assessed approach is expected to facilitate more effective execution of queries in the data mart environment. The emphasis is placed on the comparison of the query execution performance using database schemas, each built using traditional and the experimental approach. The tests are done in the environment of selected modern Business Intelligence tools, and using two test queries with varying output dataset sizes. The results show that the use of the database schema, created according to the experimental data modeling approach, had positive impact on the querying performance in several cases. The magnitude of impact on the querying performance, however, varied depending on each query's respective resulting dataset size.
Highlights
Multidimensional data modeling principles are one of the cornerstones of the Business Intelligence (BI) system’s design and development process
The results show that the use of the database schema, created according to the experimental data modeling approach, had positive impact on the querying performance in several cases
Results for the query Q2 (Fig. 3) show that the usage of the anchor data modeling (ADM) based database schema was beneficial in 2 cases
Summary
Multidimensional data modeling principles are one of the cornerstones of the Business Intelligence (BI) system’s design and development process. These principles were introduced more than 3 decades ago by Ralph Kimball and developed to today’s well-known bus architecture and dimensional modeling methodology [1]. The business process management will always rely on the analysis of historical facts in relation to current real-time data. This is supported in [2] where the authors state that the data and information integration are the most fundamental issues in the integration of decision support systems into processes, to enhance decision support performance
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call