超对等网络中的轮廓查询优化

发布时间：2018-05-09 21:13

本文选题：轮廓查询 + SUPER-PEER体系架构　；参考：《电子学报》2013年08期

【摘要】：轮廓查询是近年来信息服务领域的一个研究重点和热点.现有的三阶段算法TPAOSS(Three-PhaseAlgo-rithm for Optimizing Skyline Scalar)至少存在如下两个缺陷:(1)在TPAOSS算法的第3阶段中,当网络节点上的对象个数较多时,Bloom filter的长度将呈指数级增长,从而严重影响获取子空间重复值的效率以及占用内存空间的大小;(2)TPAOSS算法只考虑预处理阶段的时间代价,而没有考虑各网络节点进行局部或全局子空间轮廓查询计算的效率.为此,提出一种适合超对等网络(Super-Peer Architecture,SPA)的子空间轮廓查询方法EPSSQDN(Efficient Processing of Sub-space Skyline Queries in Distributed Networks).EPSSQDN算法有效解决了TPAOSS算法的的两个主要性能问题,并且显著提高了SPA网络中的子空间轮廓查询处理的效率.此外,为了能够进一步降低子空间上轮廓查询的时间开销以及网络节点间的数据传输量,我们给出新颖且有效的优化策略.实验结果表明,EPSSQDN算法比TPAOSS算法更能够缩短SPA网络中子空间轮廓查询的时间开销.
[Abstract]:Contour query is a research focus and hotspot in the field of information service in recent years. In the third stage of the TPAOSS algorithm, when the number of objects on the network node is large, the length of filter will increase exponentially. Therefore, the efficiency of obtaining repeated values in subspace and the time cost of preprocessing stage are only considered in TPAOSS algorithm, and the efficiency of local or global subspace contour query is not considered. Therefore, a subspace contour query method, EPSSQDN(Efficient Processing of Sub-space Skyline Queries in Distributed Networks).EPSSQDN, which is suitable for super-peer-to-peer network super-Peer Architecture spa, is proposed to effectively solve the two main performance problems of the TPAOSS algorithm. Moreover, the efficiency of subspace contour query processing in SPA network is improved significantly. In addition, in order to further reduce the time cost of contour query on subspace and the amount of data transmission between network nodes, we propose a novel and effective optimization strategy. Experimental results show that EPS SQDN algorithm can shorten the time cost of subspace contour query in SPA network more than TPAOSS algorithm.
【作者单位】：同济大学计算机科学与工程系;同济大学嵌入式系统与服务计算教育部重点实验室;北京大学软件与微电子学院;
【基金】：国家自然科学基金(No.61272268,No.71171148) 教育部新世纪优秀人才支持计划(No.NCET-12-0413) 同济大学中央高校基本科研业务费专项资金江苏省自然科学基金(No.BK2010139)
【分类号】：TP393.02

【参考文献】