【作者】 付戈； 时杰； 李专； 张国宾； 胡红清； 朱虹；

【Author】 FU Ge~(1+),SHI Jie~2,LI Zhuan~(1,2),ZHANG Guobin~1,HU Hongqing~1,ZHU Hong~1 1.College of Computer Science,Huazhong University of Science and Technology,Wuhan 430074,China 2.Dameng Database Corporation,Wuhan 430073,China

【机构】 华中科技大学计算机科学与技术学院； 达梦数据库有限公司；

【摘要】 数据库可生存性要求在发现受损数据时采用受损限制技术,以尽快隔离受损数据从而防止由于数据共享造成受损范围进一步扩大带来的危害。在DBMS内部设计并实现了一种受损数据隔离策略语言,并提出了一种有效的受损数据隔离算法。该算法使用SQL语句动态替换技术最大限度保证隔离阶段用户请求数据的完整性,并通过子查询动态视图替换法保证数据隔离的有效性。基于TPC-W基准的吞吐率测试证明,提出的数据隔离算法具有较高的运行时效率。更多还原

【Abstract】 The survivability requires the database system to have an ability to isolate corrupted data after an intrusion activity is detected,so as to avoid damage spreading due to share of corrupted data.This paper defines and implements an isolation policy language inside DBMS(data base management system) and proposes an innovative data isolation algorithm.This algorithm guarantees users data integrity by means of substituting table with dynamic view during the phase of isolation.By substitution of all tables with dynamic views in sub-queries,the algorithm achieves a highly effectiveness.Experiments based on TPC-W benchmark show that the data isolation algorithm has a low running performance overhead.更多还原