混合数据库中的脏数据隔离优化技术仿真

Simulation of Optimization Technology of Dirty Data Isolation in Hybrid Database

研究分布式混合数据库中脏数据有效隔离的问题。分布式混合数据库中,不同区域的数据属性不同,属性体现出的威胁程度由于属性和所在区域的差异,会形成较大反差。传统的安全体系无法对数据库中的不同区域的脏数据根据属性进行高效的隔离,导致数据库中受损数据量不断增加。提出一种基于阀值和隔离标识的脏数据隔离方法,分析分布式混合数据库的实时事务模型,通过数据干扰关系给出损坏数据的确定方法,采用阀值对分布式混合数据库中的脏数据进行动态分区聚类,采用基于隔离标示向量的脏数据实时隔离算法对分布式混合数据库中的脏数据进行准确的隔离和修正,确保分布式混合数据库的完整性和准确性。实验结果说明,上述隔离方法能够对数据库中的脏数据进行准确的隔离。

In this paper, the effective isolation problem of dirty data in distributed hybrid database was studied. This paper presented a method for the isolation of dirty data based on threshold and isolation identity. Firstly, the real time transaction model in distributed hybrid database waaas analyzed. Using data interference relationship gave the determining method corrupted data. Then, the threshold value was used to make dynamical partition clustering of the dirty data in distributed hybrid database. Meanwhile, the dirty data isolation algorithm in real time based on identi fied vector was applied for dirty data accurate isolation and correction in the distributed hybrid database to ensure the completeness and accuracy of distributed hybrid database. Experimental results show that the method can accurately isolate dirty data in the database.