Clustering Model of P2P CDN Based on The Prediction of User Requirements
|
Title | Clustering Model of P2P CDN Based on The Prediction of User Requirements |
Authors | |
Abstract | With the increasing scale of the P2P (Peer to Peer) network, the contradictions among the scope expanding, the time expending reducing and the network loading of content search is growing more serious. As the required contents and replicas are distributed in the P2P network, the searching scope reduction will result in achieving less or even no required replicas. Contrarily, expanding the searching scope will increasing the search time and network loading. In this paper, an clustering model of P2P content distribution network (CDN) based on the prediction of user requirements is proposed in which the organization of the P2P CDN is clustered continuously according to the improvement of the prediction of user requirements. In the model, the peers and the surrogates that have the requiring contents of a designate peer will be assembled at nearby locations to the designated peer, and then more replicas can be found in smaller search scope and shorter time expending. Simulative experiments indicate the clustering model can acquire much smaller mean response time (MRT) and mean distribution time (MDT). |
Publisher | ACADEMY PUBLISHER |
Date | 2012-03-01 |
Source | Journal of Networks Vol 7, No 3 (2012): Special Issue: Selected Papers of the 13th International Conference on Computer |
Rights | Copyright © ACADEMY PUBLISHER - All Rights Reserved.To request permission, please check out URL: http://www.academypublisher.com/copyrightpermission.html. |