内容寻址网络性能优化硕士学位课题论文.docVIP

硕士学位论文 论文题目 内容寻址网络性能的优化 及内容的有效定位 研究生姓名 齐庆虎 导师姓名 李津生 教授 专业 通信与信息系统 论文完成时间 2005年5月 摘 要 目前互联网主要技术模式是Client/Server(C/S)，即客户服务器模式。但是随着网络规模的扩大，尤其是Web技术的发展使得许多Web服务器成为信息的主要提供源，也就使得整个Internet系统依附于这些少量的服务器节点，从而限定了系统的应用模式，也引发了一些问题：服务器的负担太重，难以管理大量的客户机，系统的性能容易变坏；当流量增加时，容易在服务器处产生瓶颈；在服务器端对应用程序的微小修改，都有可能导致所有客户端程序的重新安装；并且只有少数服务器被充分的利用，而其他很多机器上的资源都被浪费了。 1999年，第一种P2P(Peer to Peer)文件共享服务Napster出现了。Napster允许每个终端用户（称为Peer）Nowadays, the main technology mode of the internet is Client/Server.Whereas, some problems have appeared with the increase of the network scale.To resolve these problems,P2P has appeared. In 1999,the first “Peer_to_Peer”or P2P file-sharing applicationNapster was launched.Napster allowed individual end-users(called peers) to share the MP3 music stored on their local computers directly with one another over the internet. Within one year,Napster had grown to a user population of over 50 million users.After three years,despite the closure of Napster,the phenomenon of file-sharing continued its dramatic growth. The P2P architecture with its use of low-cost,grass-roots resoures and its decentralized nature that does not rely on any form of centrally managed infrastructure，represents a significant departure from the C/S architecture.These unique characteristics,allow P2P systems to support the rapid and low-cost deployment of powerful large-scale applications that would not be possible with the current C/S architecture. There are two key pieces to a P2P system:the lookup mechanism used to locate the desired files and the actual files download.The decentralized storage in P2P systems makes the file transfer process scalable,but the trouble is how to find the peer which has the desired files.To resolve this problem is the emphases of this paper.I.e,given a file identifier key ,how can we find the peer holding the desired files?Ideally,a solution to this problem must be scalable,must find files quickly,and must be resilient to the frequent arrival and depart