| With the development of computer technology and the gradual deepening of network usage in financial system, the computer information system has been widely used in financial areas. Database is the storage center for critical data and plays an important role in financial system, and it is the most attractive goal for the attackers. Besides, all kinds of computer crimes against financial databases come one after another. In large financial system, there are a large number of databases with high capacity. The current security mechanism for single database can't meet the requirement of this, so we have to focus on the security management for large financial database systems.Access control is the first line of defense in database security and it is precondition for database authentication and authorization. Audit plays an important role in high security-level system, which provides evidence for later follow-ups. Attack detection is an active and positive safety mechanism, it can prevent unauthorized access and malicious acts by monitoring database status and user activities.In this paper, I research access control, security audit and attack detection for large financial database systems. First, describe the application-role based access control model, and it can achieve unified security management for large number of database systems. Secondly, design and realize a central based SQL Server database audit system, which offers users a choice to select audit trace events. Besides, the system holds audit data in central server by partition tables. It not only ensures the security of audit records, but also reduces the performance impact on financial application databases. Finally, I describe the attack detection, including trigger-based real-time misuse detection and SQL-based association rule mining for anomaly detection. In all, we can achieve the security database management for large financial database system by combination of these three security mechanisms. |
PDF Full Text Request