空间数据库的索引技术

由于空间数据库中的数据量很大,因此空间数据库查询的开销一般要比关系数据库大,特别是查询语句的条件谓词中包含一些对空间数据操作的函数,计算这些函数的开销远比数值或字符串的比较要大。如果用顺序扫描的方法查询,则效率非常低。因此,为了提高查询效率,采用空间索引是十分必要的。目前人们的研究工作更多地集中在空间数据的多维索引的研究上。全面地总结了当前空间数据库领域中空间索引的研究进展,然后介绍了目前空间数据库中广为采用且比较新的4种索引方法:(1)R树(2)K-D树(3)Quad树(4)GiST。最后指出在空间数据库中的高维索引的研究是目前前沿研究的热点。

Indexing techniques in spatial databases

Due to a great quantity of data in spatial databases, so in general the cost of query in spatial databases is higher than in relational databases. Especially, when there are some functions, in predicate of query, which deal with spatial data, the cost of computing these functions is higher than the cost of comparing strings and numerical values. If query strategy is to scan sequentially spatial data, then the efficiency will be very low. For improving query efficiency, it is necessary to adopt spatial indexing techniques. Indexing techniques in spatial databases have gradually caused the attentions of many people. Research advance of indexing techniques for spatial databases is summarized, and then four new and often used indexing methods, including R-tree, K-D-tree, Quad tree and Generalized search tree, are introduced. Finally, it is pointed out that the high dimensional index is a hot research field in spatial databases.