数据流上周期更新滑动窗口的连接算法  

Periodically updating sliding window join algorithms over data streams

作　　者：王伟平[1] 李建中[1] 张冬冬[1] 郭龙江[1] 

机构地区：[1]哈尔滨工业大学计算机科学与技术学院,黑龙江哈尔滨150001

出　　处：《哈尔滨工业大学学报》2005年第6期756-759,785,共5页Journal of Harbin Institute of Technology

基　　金：国家自然科学基金项目资助.

摘　　要：根据滑动窗口的更新粒度定义了两类滑动窗口:连续更新滑动窗口和周期更新滑动窗口.由于已有的滑动窗口上的查询处理算法都是针对连续更新滑动窗口提出的,并不适用于周期更新的滑动窗口,因此提出了三种有效的周期更新滑动窗口连接算法,即BSHJ、BSNLJ和BSNHJ算法.理论分析和试验结果表明BSNHJ算法具有最好的性能.According to the updating granularity, two kinds of sliding window are defined: continuously updating sliding window and periodically updating sliding window. All the existed sliding window algorithms over data streams are designed for continuously updating sliding window, which are not suitable for periodically updating sliding window. To address this problem, three effective join algorithms for periodically updated sliding window, BSHJ, BSNLJ and BSNHJ, are proposed. Theoretical analysis and experimental results show that BSNHJ achieves optimal performance.

关 键 词：数据流 周期更新滑动窗口 连接算法 

分 类 号：TP311.132[自动化与计算机技术—计算机软件与理论]