摘要
现有P2P网络规模大、动态性高、异构性强,有效的搜索技术一直是P2P系统研究中的核心问题。本文针对无结构P2P网络泛洪搜索机制的盲目性所导致的查询开销大、效率低的问题,提出了一种基于语义相似的P2P搜索机制SRVN,通过积累历史搜索经验获得路由指引信息,帮助搜索结点快速发现与搜索内容最相关的结点,从而提高搜索效率和目标命中率。实验结果显示,使用SRVN搜索机制,有效地提高了Gnutella查询性能。
Due to the increasing size of P2P systems and their characteristic of being dynamic and heterogeneous, effective searching techniques have become an important research issue and received much attention recently. On studying the key facts in unstructured P2P searching technologies and analyzing their limitations, this thesis proposes a similari- ty-based semantics searching technology. It gets routing information according to searching history to help recover the target which is most related to the searching content, thus improving the efficiency of searching and the hit rate. Simulation results prove the effectiveness and efficiency of the proposed searching mechanism.
出处
《计算机科学》
CSCD
北大核心
2007年第5期52-56,共5页
Computer Science