阈值上限一定时超大规模数据库的查询方法研究

在确定查询阈值上限时,当前数据库查询方法只能使用查询树的一条路径,并行性能较差;在对超大规模数据进行查询时,具有查询时间长、响应不及时的弊端。为此,提出一种新的阈值上限一定时超大规模数据库的查询方法,通过RSA算法对超大规模数据库进行加密和解密处理,依据多叉树的定义构造多叉索引树,在此基础上计算各代表数据点与查询点之间的距离;在阈值上限一定时,利用多个从节点机并行处理后,使用主节点机汇集结果,运用后续子树递归处理,直至获取查询结果。实验结果表明,所提方法不仅具有很高的查询效率,而且CPU使用率很高。

Threshold limit must study on the method of large scale database query#$NL Huang Yu

In the query threshold limit is determined, a path of the database query method can only use the query tree, poor performance in parallel, queries on large scale data, has disadvantages of long time, query response is not timely. For this, put forward a method of query threshold limit of new large scale database when, through the RSA algorithm for encryption and decryption processing on large scale database, according to the definition of binary index tree structure tree, calculated between each representing a data point to the query point distance on the basis of this, the threshold limit. The use of multiple parallel processing nodes from the machine, use the main node collection results, using the following recursive subtree, to obtain query results. The experimental results show that the proposed method not only has high query efficiency, but also has high usage rate of CPU.