Study And Design On Security Of Data Warehouse In SZDSS

Analyzing large amount of data collected through daily operation is a imminent task, and so data warehouse technology is developing quickly. Because the data in data warehouses is clear, integrated and valuable, it is an attractive target for hacker. So the security of data warehouse is a deserved research subject in today's information and network security field.The paper analyzes some security problems of data warehouse in a decision assistant system SZDSS and provide several settlement methods. SZDSS integrates some existed management information system, providing decision support function through data warehouse and OLAP technology.In this paper we discuss a new access control model for data warehouses and OLAP which named OlapACL offering greater expressiveness and highly increased usability with respect to security. Based on a simple formal notation we will show that access privileges can be expressed more intuitively than using SQL's grant statements in data warehouses. It is hierarchical, simple, strict, convenient and efficient.To store data safely, the data in data warehouse must be encrypted. But database encryption is much different from general information encryption, and so we present some solutions such as random encryption and improved Hill encryption which are more adaptable for data warehouse encryption. We also use some existed techniques to settle other security problems. We use firewall and VPN to improve the inner system's security. Data transmission security is another problem which we can adopt SSL to solve.Finally we sum up our research works and points out the further research work.