键值数据库中弱数据一致性改进技术研究

当前数据库中弱数据一致性维护方法存在读写性能低、集群性能差的问题,为此,提出一种键值数据库中弱数据一致性改进方法。采用数学归纳法建立hash值空间的上下限,利用哈希函数对数据库中的冗余数据进行优化；在此基础之上,通过建立未知弱数据和己知弱数据之间的回归预测模型,同时利用BP网络算法对数据迁移中的缺失数据进行采集,以保证数据的一致性。实验结果显示,与传统方法相比,采用本文方法读写性能更好,且对数据块越大,吞吐量成倍提升。

Study on the key technology to improve the consistency of the data in the database

In the current database data consistency maintenance method has read and write performance is low, the problem of poor performance of cluster, therefore, proposed an improved key database in data consistency. Bound by mathematical induction method to establish the hash value space, the redundant data in the database is optimized using hash function; on this basis, through the establishment of the data of known and unknown weak weak regression between the data prediction model, and using the BP network algorithm to collect the missing data in the data transfer, in order to ensure the consistency of the data. The experimental results show that compared with the traditional method, this method has better performance in reading and writing, and a block of data The larger the throughput is multiplied.