In order to handle massive spatial data quickly and efficiently, a superior solution is to store and handle them in parallel spatial database management systems under the environment of PC cluster at present, and thus...
详细信息
ISBN:
(纸本)9780819465290
In order to handle massive spatial data quickly and efficiently, a superior solution is to store and handle them in parallel spatial database management systems under the environment of PC cluster at present, and thus its spatial partitioning strategy of data needs solving first. hilbert spatial ordering code based on hilbert space-filling curve is an excellent linear mapping method, and gets wider and wider applications in processing spatial data. After studying hilbert curve, this paper proposes a new and efficient algorithm for the generation of hilbertcode, and it has overcome drawbacks of the traditional algorithm. Then hilbertcode is applied to spatial partitioning with the method of cluster analysis, and a concrete method is given, which fully considers characteristics of spatial data, such as the aggregation of spatial data, reduces the time of disks accesses, and achieves better performance by experiments than the compulsory partitioning of ORACLE spatial based on X coordinate values and (or) Y coordinate values in subsequent parallel processing of spatial data.
暂无评论