期刊文献+

一种新的服务器部署及其关键技术

New Server Placement Strategy and it Key Technology
下载PDF
导出
摘要 提出一种新的基于网络坐标的服务器部署方案来应对传统服务器部署方案存的先验知识不足与搜索效率不高的缺陷.该方案通过构建网络坐标的方式来获取网络拓扑信息.与传统网络测量相比,它只需少量的测量成本即可获得较完整的网络拓扑信息.在此基础上,引入分层聚类算法来实现网络坐标样本点的聚类并获得部署方案.该算法简单易实现,且能避免传统服务器部署模型的NP-hard问题.实验结果表明,基于网络坐标的服务器部署方案在取得全局优化的同时,还具有测量成本低、表现直观等优点. This paper focuses on how to deal with the disadvantages of traditional server placement strategies, such as incomplete In- ternet topology information getting and low researching effective. To this end, the paper proposes a new server placement strategy, which gets the Intemet topology information with the network coordinates. Comparing with traditional measurement, the network co- ordinate can get the Internet topology information with less measurement overhead. Then, a modified hierarchical clustering algorithm is introduced, which can avoid the NP-hard problem of the traditional server placement strategies. The results of experiment show that our server placement strategy can achieve the global internet optimum, with the advantages of less measurement overhead and per- formance intuitive.
出处 《小型微型计算机系统》 CSCD 北大核心 2012年第9期1987-1991,共5页 Journal of Chinese Computer Systems
关键词 服务器部署 网络坐标 分层聚类 坐标基准点 server placement the network coordinate hierarchical clustering landmark
  • 相关文献

参考文献10

  • 1Charikar M,Guha S, et. al. A constant-factor approximation algorithm for the k-median problem [ C ]. Proc. Ann. ACM Symp. Theory of Computing, 1999 : 1-10.
  • 2Jain K, Vazirani V V. Approximation algorithms for metric facility location and k-median problems using the primal-dual schema and lagrangian relaxation[ M]. Journal of the ACM, 2001:274-296.
  • 3Cameron C W, Low S H, Wei D X. High-density model for server allocation and placement [ C ]. Proc. ACM SIGMETRICS, 2002, 30(1) :152-159.
  • 4Ng T S E, Zhang H. Predicting internet network distance with coordinates-based approaches[ C]. Proc. of IEEE INFOCOM,2002.
  • 5D' Andrade R G. U-statistic hierarchical clustering [ M ]. Psychometrika, 1978,43 ( 1 ) :59-67, 1978.
  • 6Bocker A, Derksen S, Schmidt E, et al. Hierarchical k-means clustering [ EB/OL]. http://gecco, org. chemie, uni-frankfurt, de/hkmeans/H-k-means, pdf, 2011-034)1.
  • 7Francis P, Jamin S, Paxson V, et al. An architecture for a global Internet host distance estimation service[ C]. Proceedings of IEEE INFOCOM '99, 1999.
  • 8Wen Z, Rabinovich M. Network distance estimation with dynamic landmark triangles[ C]. Proceedings of SIGMETRICS '08,2008.
  • 9Arthur D, Vassilvitskii S. How slow is the k-means method [ C ]. Proceedings of the 22nd Symposium on Computational Geometry (SoCG), 2006.
  • 10Frahling G, Sohler C. A fast k-means implementation using coresets[C]. Proceedings of the Twenty-second Annual Symposium on Computational Geometry (SoCG) ,2006.

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部