Research On OLAP Storage Structure Based On Dimensional Hierarchical Clustering

In the research area of data warehouse (DW) and decision support system (DSS) online analytical processing (OLAP) is a hot issue .In general the OLAP queries on the mass data set are complicated, a quick query response is needed. So how to improve the data querying and updating efficiency is an important issue in data warehouse.After detailed analyses and study of the OLAP theory and application at both home and abroad, this dissertation makes some research of OLAP technology as follows:Firstly, this dissertation analyzes the theory of OLAP and discusses its basic conception, structure, the way of data organization and the basic demand an OLAP system needed.Secondly, this dissertation analyzes the now existing data cube storage structure's characteristic, the way of implement. In some degree these storage structure improve the query performance, but they still cannot meet the need of quick query and update in practice. Recently, in ROLAP system the application of hierarchical clustering theory is relatively mature and gain achievement, improving the performance of ROLAP system. This dissertation try to apply the hierarchical clustering theory into MOLAP storage structure, and constract the Dimensional Hierarchical clustered cube (DHC-Cube).This dissertation compares the dimensional hierarchical clustered cube with the existing cube storage structure .Analysis demonstrates that DHC-Cube Structure is efficient and able to minimize data updating costs and enhance the query speed.Finally, Using the relative theory of dimensional hierarchical clustering into the project and gain certain achievement.