Research And Design Of Data Warehouse System In Secutities Company

Since the early 1990s,with the development of data warehouse and the maturity of the this technology, The financial institutions both at home and abroad begun using data warehouse technology to do business analysis, profits analysis and customer analysis, and got good profit. In the CRM, the financial institutions have established the idea of "customers first" and widely applied the technology of data warehouse, This makes domestic and foreign financial institutions, particularly investment banks, commercial banks and securities companies, will get an advantage in the fierce competitive environment.Firstly, this thesis deeply analyses the situation of fierce competition which domestic securities companies are facing at and the implement of data warehouse in domestic and foreign financial institutions, and discusses the status of data warehouse related technology and there development prospect. Based on widely mastering the knowledge of security industry, and full analysis of a domestic security company's business and development objectives, this thesis designs a data warehouse for the CRM system of this company. There are a large amount of data in the business database, so the system uses distributional and the three-level structure.Secondly, based on the objective of CRM system for the customer's analysis of bank, the thesis designs the data model of this data warehouse system. The thesis confirms four subjects and granularity of this system, including the customer, the account, the transaction and the security. After analyzing these subjects, the thesis designs the fact table, dimension tables, index strategy, storage structure, storage strategy etc.Finally, the thesis makes a key research on ETL, which is a key technology in building a data warehouse system. Because there are a large mount of data and some heterogenous data in the company, it proposes the system data extraction strategy, data mapping relation, the data cleaning and loading and so on, and realizes parts of ETL process.