R 树， 存取方式的当中最普遍的的用长方形， 是根据区域的启发式优化围绕的长方形在各个内在结点通过运行在一张规范化的试验平台下高度变化的数据， 询问和操作众多的实验， 我们能设计合并联合的优化的R* 树区域， 各个附寄的长方形边际和交叠在目录使用我们规范化的试验床在详尽表现比较，它结果R* 树清楚地胜过现有的R 树变形Guttman’s线性和二次方R 树和R 树的格林变形，R* 树的这优势举行为不同的型询问和操作， 譬如地图覆盖物。 为两个长方形和多维点在所有实验从一个实用观点R* 树是非常有吸引力的由于以下二个原因1 它高效率地支持点和空间数据同时和2 它实施费用比那少许高级其它R 树。-R tree, which forms of access to the most common use of a rectangular shape, which is based on heuristic optimization rectangular inherent in all nodes running on a standardized test platform height data, inquiries and the operation of many experiments, we can design the optimal merging of the joint R* Tree region, judging from the various and overlapping rectangular marginal in the directory using our standardized test bed in a detailed performance comparisons, and it results R* tree clearly better than the existing R-tree deformation Guttman's linear and quadratic R-tree and R Green Tree deformation, R* Tree of this advantage at different type of inquiry and operation, such as maps cover. For two and multi-dimensional rectangular in all experiments from a practical point of R* tr