Data Warehouse And Olap Technology, Research And Applications

With the development of database technology and the expanding of database scale, people hope to refine the useful information from existing data to serve for decision; therefore, decision support system emerges as the time require. There have been growing interests in the techniques of data warehouse, On-line Analytical Processing and data mining since they support the decision-making.OLAP organizes multidimensional data against data cube model, which is convenient to queries. OLAP usually involves complex queries on very large multidimensional datasets. Running analytical queries directly against the huge raw data volume of a data warehouse results in unacceptable query performance. The solution to this problem is storing materialized views in the warehouse, which pre-aggregate the data and thus avoid raw data access and speed up queries. This thesis works on an important aspect of physical designing of data warehouse: the selection of materialized views. Algorithms to be used while building the data warehouse are presented.This thesis studies a greedy algorithm of views selection .we divide the algorithm into a views path selection phase and a views computation phase and select views based on searching view in the layer of data cube. So the algorithm complexity decreases from O(k.n ) to O (k.n.logn). The analysis and experimental study show that the algorithm has better solution and high efficiency.Then, this thesis puts out an application illustration in a sale system, designs and implements an OLAP system prototype based on MS SQL Server Analysis Services. The multidimensional data model of the sample database is designed as star schema. The techniques of establishment and access for multidimensional dataset are discussed. We implement multidimensional queries by the technique of MDX, ADO MD, which realizes a basis for the analysis of commercial data and supplies strategic decisions for the decision-maker.