论文部分内容阅读
针对空间信息的非均匀分布和邻近性特点,以及海量空间信息处理中逻辑覆盖网络与物理网络不一致的问题,引入对等网络(P2P)分层理论和非均匀Hilbert曲线,提出一种适合空间信息处理的P2P分层网络模型(SIPLNM).该模型分为两层:超级节点层和区域节点层,超级节点层是由负责相应区域的超级节点组成的,区域节点层由一个区域内的所有节点组成.通过非均匀Hilbert曲线保持空间对象之间的邻近性,实现空间信息在划分区域之间的均匀分布.区域内采用hash空间信息主题方式,实现空间信息在第二层节点的均衡分布.实验表明,本方法能够有效地克服现有区域划分和空间信息分布方法的不足,在SIPLNM各节点中,均有良好的分布均衡性.
Aiming at the non-uniform distribution and proximity of spatial information and the inconsistency between logical overlay network and physical network in massive spatial information processing, this paper introduces P2P hierarchical theory and non-uniform Hilbert curve, (SIPLNM) .This model is divided into two layers: the super node layer and the regional node layer, the super node layer is composed of the super node responsible for the corresponding area, and the regional node layer is composed of all nodes in one area The non-uniform Hilbert curve is used to keep the spatial object’s proximity and to achieve the uniform distribution of spatial information between the divided regions.The hash space information subject is adopted in the region to achieve the balanced distribution of spatial information at the second node.Experiment It shows that this method can effectively overcome the shortcomings of existing regional and spatial information distribution methods and has good balance among all nodes in SIPLNM.