期刊名称：International Journal of Grid and Distributed Computing
印刷版ISSN：2005-4262
出版年度：2016
卷号：9
期号：7
页码：159-168
DOI：10.14257/ijgdc.2016.9.7.17
出版社：SERSC
摘要：Social network has become the important platform for the current social individual to exchange information and access various media. Traditional searching and locating method is faced with the characteristics of the high order correlation and implicit correlation for large-scale users, and user-association multidimensional and heterogeneity. In order to effectively deal with large-scale social network data, and improve the efficiency of user's locating, this article introduces the peer-to-peer distributed searching mechanism with the help of the cloud computing platform. This searching mechanism assigns user a logical identifier, and matches the underlying physical address and the upper users' logical address, so as to build the cloud logical topology structure of social network. This paper designs a K neighbor discovery algorithm. It is used to cluster the nodes according to the features of the user, so as to realize the quickly locating of social network. The performance of the algorithm is analyzed according to the user's searching logic path length of social network, and information amount of routing state. The simulation of the algorithm is evaluated by maintenance costs of average network aggregation coverage and query time. The performance analysis and simulation results demonstrate that social network cloud has good performance and searching efficiency.
