节点文献
基于P2P的校园网搜索引擎的设计与实现
Design and Implementation of University Search Engine Based on P2P
【作者】 温丽明;
【导师】 王艳清;
【作者基本信息】 北京化工大学 , 计算机应用技术, 2008, 硕士
【摘要】 校园网是学校信息化建设的基础设施,是广大师生完成教学和科研的一个重要平台。基于P2P校园网搜索引擎利用P2P分布式体系结构,将网络中的资源分布在互联网中的各个节点上。并组建了基于“小世界”现象的兴趣网络,把整个资源共享范围缩小到校园网内,有效解决了P2P网络的带宽占用问题,减轻了校园网路由出口的压力,同时也提高了节点间彼此传输文件的速度。基于P2P校园网搜索引擎采用了现今流行的两种开源技术——JXTA和Nutch。利用JXTA建立一个通用的、统一的P2P网络平台,采用IP多播技术的动态发现机制和汇聚对等体的静态发现机制,减少了资源浪费。同时采用XML方式,对其资源信息进行描述及传递,并采用了XML信息的缓存技术,通告在其经过的所有中间对等体处都会被缓存,从而使系统性能随着系统“变老”而不断提升。利用Nutch技术构建了基于Lucene的Web搜索引擎,比目前普遍采用的搜索引擎(如Baidu、google等)更具有针对性,将搜索范围限制在校园主页及其相关链接中。在网页爬行过程中采用了广度优先遍历方式,并允许服务器管理者自由指定搜索的深度,使该系统更具有实效性。
【Abstract】 University network is the basic facility on the construction of school informationlize. It is also an important platform, with which the teachers and students will teach and research. The university network based on P2P save its resource in the various peers with P2P distributed architecture. Besides, the program builds the interest network based on "small word", which means that the scope of the files shared is limited within the university network. So we can make full use of the resource in our university. It also provides an effective solution to the bandwidth occupied in order to reduce the big pressure on our network export. At the same time, it speeds up file transfers between peers.The implementation of university search engine based on P2P using two popular open-source technology - JXTA and Nutch. On one hand, the system builds a common and unified P2P network with JXTA. JXTA provides the dynamic discovery mechanism with ip multicast and the static discovery mechanism with rendezvous peers. It can reduce the waste of the bandwidth. At the same time, it describes and transfers information with XML. The advertisement is saved in the middle peers, which just transfer information. On the other hand, the system builds web search engine based on Lucene technology with Nutch. The aim of designing it is more specific than the most popular search engines (such as Baidu, google, etc.). It just search the university home page and the related links. Web crawling use the breadth-first traversal and allow the server administrator to decide the depth of search, so the system is more effective.
- 【网络出版投稿人】 北京化工大学 【网络出版年期】2008年 11期
- 【分类号】TP393.18;TP391.3
- 【被引频次】2
- 【下载频次】407