基于多层Bayesian信任网的P2P负载均衡模型

doi:10.3785/j.issn.1008-973X.2010.09.008

2010, Vol. 44

Issue (9): 1676-1680 DOI: 10.3785/j.issn.1008-973X.2010.09.008

自动化技术、计算机技术

基于多层Bayesian信任网的P2P负载均衡模型

宋广华1,2,夏莹杰2,郑耀1,2,毛小云2

1.浙江大学航空航天学院,浙江杭州 310027； 2. 浙江大学计算机科学与技术学院,浙江杭州 310027

P2P loadbalance model based on multi-layer Bayesian trust network

SONG Guang-hua1,2, XIA Ying-jie2, ZHENG Yao1,2, MAO Xiao-yun2

1.College of Aeronautics and Astronautics, Zhejiang University, Hangzhou 310027, China;
2. College of Computer Science and Technology, Zhejiang University, Hangzhou 310027, China

全文: PDF HTML

摘要：

针对基于信誉机制的P2P（PeertoPeer）网络中的负载不均衡现象,提出一种基于多层Bayesian信任网的P2P负载均衡模型,介绍了如何在Bayesian信任网中引入多种负载指标以及它们之间的各种关联,并且使用该结构预测P2P网络中各服务节点的负载,使整个网络达到负载均衡.对包含负载指标的多层Bayesian信任模型和动态负载模型进行了实验比较.实验结果表明：基于Bayesian信任网的P2P网络负载均衡模型运用节点的历史交互信息,结合Bayesian预测理论,影响资源请求节点选择服务节点,实现P2P负载的有效均衡,确保整个P2P网络资源利用的高效性、节点服务的可靠性以及整个网络的稳定性.

Abstract:

A PeertoPeer(P2P) load balance model based on multilayer Bayesian trust network was designed to deal with the load balance problems in the trustbased P2P network. Some load indexes and their relations were added to the model to predict the workload of the peers. Experiments were done for the presented model and the dynamic load model, respectively. The results show that, by exploiting the interactions history and the Bayesian prediction theories, the load balance model based on multilayer Bayesian trust network influences the peer selection process and can effectively balance the load of the P2P network, thereafter improving the resource utilization ratio, the quality and reliability of the peers services．

出版日期: 2010-09-01

TP 393.03

基金资助:

浙江省科技计划国际合作重点资助项目.

作者简介: 宋广华1,2,夏莹杰2,郑耀1,2,毛小云2

	服务
	把本文推荐给朋友
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章

引用本文:

宋广华, 夏莹杰, 郑耀, 毛小云. 基于多层Bayesian信任网的P2P负载均衡模型[J]. J4, 2010, 44(9): 1676-1680.

SONG An-Hua, JIA Ying-Jie, ZHENG Yao, MAO Xiao-Yun. P2P loadbalance model based on multi-layer Bayesian trust network. J4, 2010, 44(9): 1676-1680.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2010.09.008 或 http://www.zjujournals.com/eng/CN/Y2010/V44/I9/1676

［1］ XIA Y, SONG G, ZHENG Y, et al. R2P: a peertopeer transfer system based on role and reputation ［C］∥ Proceedings of International Workshop on Knowledge Discovery and Data Mining. Adelaide: IEEE, 2008:136141．
［2］ KUNZ T. The influence of different workload description on a heuristic load balancing scheme ［J］. IEEE Transactions on Software Engineering, 2007, 17(7):126131．
［3］ MIRCHANDANEY R, TOWSLEY D, STANKOVIC J. Analysis of the effects of delays on load sharing ［J］. IEEE Transactions on Computers, 1989, 38 (11):15131525．
［4］ GODFREY B, LAKSHMINARAYANAN K, SURANA S. Load balancing in dynamic structured P2P systems ［C］∥ Proceedings of INFOCOM. Hong Kong: IEEE, 2004:4650．
［5］ BYERS J, CONSIDINE J, MITZENMACHER M. Simple load balancing for distributed hash tables ［C］∥ Proceedings of International Workshop on PeertoPeer Systems. Berkeley: IEEE, 2003:8087．
［6］ MEHRA P, WAH W. Automated learning of workload measures for load balancing on a distributed system ［C］∥ Proceedings of International Conference on Parallel Processing. New York: IEEE ,1993:263270．
［7］鞠九滨, 杨鳃, 徐高潮. 使用资源利用率作为负载平衡系统的负载指标［J］. 软件学报,1996, 7(4):238243．
JU Jiubin, YANG Kun, XU Gaochao. Using resource utilization as load index in dynamic load balancing ［J］. Journal of Software, 1996, 7(4):238243．
［8］夏莹杰, 宋广华, 郑耀, 等. 基于零知识证明和Bayesian信誉网的小世界P2P模型［J］. 浙江大学学报: 工学版，2010，44（1）：5660．
XIA Yingjie, SONG Guanghua, ZHENG Yao, et al. Trusted small world P2P model based on zero knowledge interactive proof and Bayesian trust network ［J］. Journal of Zhejiang University: Engineering Science ,2010，44（1）：5660．

[1]	占志峰, 邢卫, 鲁东明. 延迟变化紧密的多核心组播树快速构建算法[J]. J4, 2013, 47(1): 29-36.
[2]	夏莹杰, 宋广华, 朱明哲, 等. 基于零知识交互式证明和Bayesian信誉网的小世界P2P模型[J]. J4, 2010, 44(1): 56-60.

Viewed

Full text

Abstract

Cited

Shared

Discussed