空间数据访问规律的动态统计算法
Dynamic Statistical Algorithm of Spatial Data Access Laws
-
摘要: 根据日志信息得到的静态访问分布,不能真实反映系统当前的访问特征,而大规模分布式节点环境下,简单的复制和分发将带来不可承载的网络流量。为此,提出一种基于P2P的海量空间数据访问分布动态统计融合算法。通过节点映射和优选算法,充分利用节点闲置资源,优先选择服务能力"好"的节点进行统计信息的融合。实验表明,该算法能满足大规模节点下空间数据访问分布的动态融合要求,且效率较高。Abstract: The tile access has dynamic features(server peer capability,storage device and hot tiles have dynamic features) and the Hotmap model based on the historical log information can not reflect the real system’s current global information.Reproduction and distribution will produce a huge network flow rate in a large scale distributed nodes environment.A dynamic statistics algorithm for the distribution rule of the spatial data based on P2P is proposed to resolve above-mentioned problems.The service capabilities of the service nodes are calculated in this algorithm.The node agents with good service capabilities are chosen preferentially in the group to fuse dynamic statistical information.The experimental results show that the algorithm can meet the need of dynamic statistics in large scale distributed modes environment with high efficiency.