| 1 (2008) | |||||||||||||
Abstract | |||||||||||||
| 1 Introduction The Internet provides a platform for decentralized applications where content or services are requested, stored, and provided by very large number of loosely coupled hosts. In these applications, there is no centralized support. Requests are forwarded from peer to peer, and each peer can become a server for each item. Many supporting architectures for such applications are such that the search pattern for each request is "routed " that is, every peer has a list of neighbors, and for each request a peer might receive, there is a preferred neighbor that brings the request "closer " to its destination. Therefore, the forwarding pattern of each request forms a tree, with edges directed towards the root. We refer to such a structure as a distribution tree. | |||||||||||||
Details der Publikation | |||||||||||||
| |||||||||||||