| The multi-dimensional analysis show system which is in the background of the electronic commerce system based on Web become important constitute of the electronic commerce system. It become an important problem needed to be solved that how to design the multi-dimensional analysis display system to satisfy enterprise's practical requirements. Through the study the basic theory of data warehouse, online analytical processing (OLAP), XML technology and multidimensional expressions (MDX) syntax, this paper presents a solution for multi-dimensional analysis show system.As the speed of cube query processing in multi-dimensional analysis has direct impact on the efficiency of query of the system, in order to improve the performance, this project employs the method of materialized view and uses Mondrian as the multi-dimensional analysis server, which is an open source software. In query analysis process, Mondrian can match the materialized view and query, making the materialized view can also store the practical data. It also pre-calculates and saves results of time-consuming connection and aggregation, operating results. The data warehouse system of multi-dimensional analysis show is implemented based on the previous solution, which can achieve online analytical processing to conduct, rotation, drill down, slice, dice and other operations based on Web. The system improved the efficiency of query and integrated with office automation software seamlessly.Based on date warehouse, on-line analytical processing technology and specific business requirements, the development environment of enterprise systems is analyzed to design the data warehouse, cube are created for multi-dimensional analysis. This thesis mainly studied the query optimization of cube. In addition, through the 2D and 3D graphic presentation of the multi-dimensional analysis operation results, the system provides the support for enterprise decision makers. |
PDF Full Text Request