An Improved two-layer Model in the Logical Level Data Warehouse Designing

Abstract

Data warehouses are centralized repositories collected from various heterogeneous sources in a wide range of time (time-variant) for decision support systems. Data warehouses are data sources used in decision making processes by online analytical processing. The process of developing a data warehouse is done through operational database analyzing, analytical requirement identification and finally designing in conceptual, logical and physical levels. In this paper, the design models at different levels of data warehouses are researched and the logical level models are compared and analyzed with respect to their properties. Finally, an improved model is proposed in logical level that combines two models (star and snowflake) in the form of a two-layered model. Queries response time measure is used to compare the proposed model with existing models. Experiments show that the proposed model improves the response time of queries. In fact, the response time of queries on the proposed model (two-layer) is improved 58.53 percent in average versus star model, and 96.61 percent in average versus snowflake model.

Keywords

• data warehouse in conceptual level
• data warehouse in logical level
• snowflake schema
• star schema
• data warehouse designing