一种非结构化P2P网络的资源搜索算法

本文针对非结构F2P网络中利用洪泛搜索机制来进行资源搜索的盲目性问题，提出一种利用节点经验积累和cache技术相结合来指导资源搜索的算法。在访搜索算法中，节点记录收到的查询主题和满足该主题的邻居节点，建立主题与邻居节点的对应关系表，当再次收到该主题时，节点就利用该表来指导资源搜索。为了加快搜索速度，将常见查询主题的记录放到cache中。该算法利用一个相似度函数来判断本地文档是否满足查询主题。仿真实验表明，该算法有效地改进了资源搜索的盲目性。减少查询带来的网络流量，提高了查询成功率。

A Resource-searched Algorithm of Unstructured P2P Network

This paper aiming at the blindfold problem of the unstructured peer-to-peer system which makes use of the flooding mechanism to search resource, this paper presents a new resource-searched algorithm of unstructured peer-to-peer system to guide the resource search by using the accumulated node experience and cache technology. In the algorithm, the node record the received query subject and the neighbor nodes fulfilling the subject, and then create a relation table between the subject and neighbor node. When the node receives the subject recorded again, it can guide the resource search by using the table. To speed the search, the records with common query subjects are put in cache. The algorithm uses a resemblance function to determine whether the local document fulfills the query subject. Simulation experiments demonstrate that the algorithm effectively improves the performance of searching resource, reduces the network traffic greatly, and enhances the searching success ratio.