Library Based On The Statistics Of The Data Warehouse Design And Implementation

With the quick development of the database applications, data volumes increased dramatically and users' requirements became more and more complex. For better decision support and fast response to analysis requests, operation environment and analysis environment and applications were separated, and the concepts and technologies of Data Warehouse were introduced.A data warehouse is a subject-oriented, integrated, time-variant, nonvolatile collection of data in support of management's decision-making process. In broad sense, the data warehouse is more than just data, is also the process managers that make information available enabling people to make informed decisions. The data warehouse technologies make good use of volumes of data that were accumulated in the long-term operation of corporations, which resolve the contradiction that data is abundant, while information is indigent, and also make it possible for the change of DSS from "model-driven" to "data-driven".The paper presents the history, concept and architecture of data warehouse, and the similarity and difference between database and data warehouse, as well as the concepts of metadata and data marts. It also introduces the history and development of decision support system. Then it introduces the theories and techniques of the design of data warehouse. Finally, it sets forth the application of data warehouse inthe Statistic and Analysis Subsystem of Labor Market Information System--thedesign and implementation of StatDB based on data warehouse. The works include the design of fact tables and dimension tables; the design of metadata; and the design and development of function modules of the StatDB.