期刊文献+

IP网络坐标抖动感知与慢启动抑制 被引量:4

IP-Based Network Coordinate Oscillation Awareness and Slow-Start Mitigation
下载PDF
导出
摘要 为了抑制Vivaldi网络坐标系统抖动现象并提高距离预测准确性,将Vivaldi归结为一类非线性方程组的迭代求解算法,并基于方程组的矛盾性提出了迭代因子的自适应估计问题。据此,定义单位化误差作为新的度量参数,通过对该参数的周期性测量实现坐标抖动的感知。进而对迭代过程进行逻辑分层,提出了一种基于慢启动策略的抖动抑制算法。实验证明,算法不依赖于迭代因子的人工选取,能够在不损失定位精度和收敛速度的前提下,将节点坐标抖动程度降低83.5%以上,并能与Vivaldi协议混合部署,实现坐标系统的平滑演进。 To mitigate oscillation phenomenon in Vivaldi Network Coordinate System (NCS) and improve the accuracy of Intemet distance prediction, Vivaldi is categorized as a kind of iterative method to solve non-linear equations and thus the adaptive estimation of iterative factor in Vivaldi is raised based on equations' contradictory. By defining a new metric unitization error and measuring it periodically, the awareness of network coordinate oscillation degree can be achieved in real time. Then we separate the iterative process of Vivaldi into 2 logical layers and propose a simple slow-start adaptive algorithm to estimate iterative factor automatically in order to mitigate network coordinate oscillation. The experiments and simulations show this algorithm can reduce the coordinate oscillation by no less than 83.5% without losing any velocity and accuracy, no matter what the initial value of iterative factor is. Furthermore, this algorithm can also be deployed in a hybrid environment together with raw Vivaldi to make the update of NCS smoothly.
出处 《电子科技大学学报》 EI CAS CSCD 北大核心 2012年第6期921-926,共6页 Journal of University of Electronic Science and Technology of China
基金 国家科技重大专项(2011ZX03002-002-03) 中央高校基本科研业务费(ZYGX2010J077)
关键词 收敛性 网络测量 网络坐标系统 VIVALDI convergence internet measurement network coordinate system Vivaldi
  • 相关文献

参考文献19

  • 1DABEK F, COX R, KAASHOEK F, et al. Vivaldi: a decentralized network coordinate system[C]//ACM SIGCOMM. Portland, OR, USA: ACM, 2004:15-26.
  • 2LEDLIE J, GARDNER P, SELTZER M. Network coordinates in the wild[C]//The 4th USENIX Symposium on Networked Systems Design & Implementation. Cambridge, MA, USA: USENIX, 2007.
  • 3AGARWAL S, LORCH R. Matchmaking for online games and other latency-sensitive P2P systems[C]//ACM SIGCOMM. Barcelona, Spain: [s.n.], 2009.
  • 4AGARWAL S, DUNAGAN J, JAIN N, et al. Volley: automated data placement for geo-distributed cloud services[C]//The 7th USENIX Symposium on Networked Systems Design & Implementation. San Jose, CA, USA: USENIX, 2010.
  • 5KOJIMA T, ASAHARA M, KONO K, et al, Embedding network coordinates into the heart of distributed hash tablcs[C]//IEEE Ninth International Conference on Peer-to-Peer Computing. Seattle, WA, USA: IEEE, 2009.
  • 6PATRIK M, DAN K, RADIM B, ct al. Study and performance of localization methods in IP based networks: Vivaldi algorithm[J]. Journal of Network and Computer Applications, 2011, 34(1): 351-367.
  • 7WANG Y, BURGENER D, FLORES M, ct al. Towards Street Level Client Independent IP Gcolocation[C]//The 8th USENIX Symposium on Networked Systems Design & Implementation. Boston, MA, USA: USENIX, 2011.
  • 8LUMEZANU C, BADEN R, SPRING N, et al. Triangle inequality variations in the intemct[C]//Pro.ceedings of the 9th ACM SIGCOMM conference on Imemet measurement conference. New York, NY, USA: ACM, 2009.
  • 9LEE S, ZHANG Z L, SAHU S, et al. On Suitability of euclidean embedding for host-based network coordinate systems[J]. IEEE/ACM Transactions on Networking, 2010, 18(1):27-40.
  • 10黄琼,刘熙,阳小龙,隆克平.T-Vivaldi:TIV感知的IP网络坐标系统[J].电子科技大学学报,2012,41(1):147-151. 被引量:4

二级参考文献41

  • 1NG E, ZHANG H. Predicting internet network distances with coordinate-based approaches[A]. Proc of IEEE INFOCOM'02[C]. New York, NY, USA, 2002. 170-179.
  • 2DABEK E COX R, KAASHOEK F, et al. Vivaldi: a decentralized network coordinate system[A]. Proc of ACM SIGCOMM'04[C]. Portland, OR, USA, 2004. 15-26.
  • 3DISCHINGER M, HAEBERLEN A, GUMMADI K P, et al. Characterizing residential broadband networks[A]. Proc of the 7th ACM SIGCOMM Conference on Internet Measurement[C]. San Diego, CA, USA, 2007. 306-311.
  • 4PAN H J, LOU Y H. A hierarchical network coordinate system[A].2010 International Conference on Educational and Information Technology[C]. Chongqing, China, 2010. 156-161.
  • 5CHEN Y, SUN P, FU X M, et al. Improving prediction accuracy of matrix factorization based network coordinate systems[A]. Proc of 19th ICCCN[C]. Zurich, Switzerland, 2010.1-8.
  • 6LEDLIE J, PIETZUCH P, SELTZER M. Stable and accurate network coordinates[A]. Proc of International Conference on Distributed Computing Systems[C]. Lisbon, Portugal, 2006.84-93.
  • 7阳小龙,周亮,王万新等.基于t检测模型的网络坐标系统输入时延预处理方法[P].中国:201010161784.8,2010,.
  • 8NC research group at Harvard[EB/OL]. http://www.eecs.harvard. edu/syrah/nc/.
  • 9李朝辉.数字图像处理及应用[M].北京:机械工业出版社,2004.
  • 10XING C Y, CHEN M. Research on optimizing embedding space dimension in network coordinate system[A]. Proc of 8th IEEE/ACIS International Conference on Computer and Information Science[C]. Shanghai, China, 2009. 240-245.

共引文献7

同被引文献64

  • 1LEDLIE J, GARDNER P, SELTZER M. Network coordinates in the wild[ C ]//Proc of the 4th USENIX Symposium on Networked Systems Design & Implementation. Cambridge : USENIX, 2007 : 299- 311.
  • 2AGARWAL S, LORCH R. Matchmaking for online games and other latency-sensitive P2P systems [ C ]//Proc of ACM SIGCOMM ' 09 Conference. New York : ACM Press, 2009 : 315- 326.
  • 3AGARWAL S, DUNAGAN J, JAIN N, et al. Volley: automated data placement for geo-distributed cloud services [ C ]//Proc of the 7th USENIX Symposium on Networked Systems Design & Implementation. San Jose, CA : USENIX, 2010.
  • 4CHEN Zhuo, CHEN Yang, DING Cong, et al. Pomelo: accurate and decentralized shortest-path distance estimation in social graphs[ C ]// Proc of ACM SIGCOMM' 11 Conference. New York: ACM Press, 2011:406-407.
  • 5KERMARREC A M, LEROY V,TREDAN G. Distributed social graph embedding[ C]//Proc of the 20th ACM International Conference on Information and Knowledge Management. New York: ACM Press, 2011 : 1209-1214.
  • 6DADEK F, COX R, KASSHOEK F, et al. Vivaldi : a decentralized network coordinate system [ C ]//Proc of ACM SIGCOMM' 04 Conference. New York :ACM Press ,2004 : 15-26.
  • 7SHAVITF Y,TANKEL T. Hyperbolic embedding of Intemet graph for distance estimation and overlay construction [ J]. IEEE/ACM Trans on Networking ,2008,16( 1 ) :25-36.
  • 8SHI Xiao-hui, CHEN Yang, DENG Bei-xing, et al. Network distance prediction based on network coordinate system [ C ]//Proc of the 5th International Conference on Grid and Cooperative Computing Workshops. Piscataway : IEEE Press ,2006 : 170-175.
  • 9PATRIK M, DAN K, RADIM B, et al. Study and performance of localization methods in IP based networks : Vivaldi algorithm [ J ]. Journal of Network and Computer Applications,2011,34( 1 ) :351-367.
  • 10WANG Y, BURGENER D, FLORES M,et al. Towards street level client independent IP geolocation [ C ]//Proc of the 8th USENIX Symposium on Networked Systems Design & Implementation, Boston, MA: USENIX,2011 : 1-14.

引证文献4

二级引证文献2

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