期刊文献+

基于时延的Flash Crowd控制模型 被引量:1

Flash Crowd Control Model Based on Time Delay
下载PDF
导出
摘要 提出了一种session级别的flash crowd控制策略SGAC(session-granularity admission control),将session控制粒度和request控制粒度相结合,采用请求平均返回时延作为检测和控制的依据.对session采取一旦接受就完成的策略,在实现对服务器过载控制的同时,保护用户session的完整性,并能自动调节新session的准入速率,以提高服务器利用率.采用真实HTTP Log进行模拟,结果表明,SGAC方法能够有效控制服务器过载,保护session的完整性,提高服务器利用率,降低接入端路由器计算开销,保护有价值的交易session. An adaptive session-granularity admission control(SGAC) method,which combines the session and request granularities,is proposed for flash crowd control.In SGAC,the average response delay is used to measure,detect,and control the flash crowd.Once a session is allowed to access the server,it will be served until it ends.Besides preventing a server from overloading,SGAC can protect the sessions’ integrity.By regulating the session served number adaptively,SGAC can improve the server’s utilization.The performance of SGAC with real HTTP log are evaluated,and the result show that SGAC can effectively prevent servers from overloading,protect sessions’ integrity,improve server’s utilization,reduce the request arrival rate,reduce the access router’s computing overhead,and protect valuable transaction sessions
出处 《软件学报》 EI CSCD 北大核心 2011年第11期2795-2809,共15页 Journal of Software
基金 国家自然科学基金(60703021 61070185) 国家高技术研究发展计划(863)(2007AA010501 2009AA01Z431)
关键词 FLASH crowd session粒度 过载控制 准入控制 flash crowd session-granularity overload control access control
  • 相关文献

参考文献19

  • 1http://www.matchtt.com/dp-bbsthread-369328.html.
  • 2http://ita.ee.lbl.gov/html/contrib/WorldCup.html.
  • 3Barford P, Plonka D. Characteristics of network traffic flow anomalies. In: Proc. of the 1 st ACM SIGCOMM Workshop on Intemet Measurement. San Francisco, 2001.69-73. [doi: 10.1145/505202.505211].
  • 4Chen X, Heidemann J. Flash crowd mitigation via adaptive admission control based on application-level observations. ACM Trans. on Internet Technology, 2005,5(3):532-569. [doi: 10.1145/1084772.1084776].
  • 5Clark DD, Fang WJ. Explicit allocation of best-effort packet delivery service. ACM/IEEE Trans. on Networking, 1998,6(4): 362-373. [doi: 10.1109/90.720870].
  • 6Chen F, Lambert D, Pinheiro JC. Incremental quantile estimation for massive tracking. In: Proc. of the 6th ACM KDD Int'l Conf. in Knowledge Discovery and Data Mining. Boston, 2000.516-522. [doi: 10.1145/347090.347195].
  • 7Franklin GF, Powell JD, Emami-Naeini A. Feedback Control of Dynamic Systems. 6th ed., Prentice Hall, 2010. 179-200.
  • 8Kamra A, Misra V, Nahum EM. Yaksha: A self-tuning controller for managing the performance of 3-tiered Web sites. In: Proc. of the Internet Workshop of Quality of Service. Montreal, 2004.47-56. [doi: 10.1109flWQOS.2004.1309356].
  • 9Le Q, Zhanikeev M, Tanaka Y. Methods of distinguishing flash crowds from spoofed dos attacks. In: Proc. of the 3th EURO-NGI Conf. on Next Generation Internet Design and Engineering. Trondheim, 2007.167-173. [do i: 10.1109/NGI.2007.371212].
  • 10Xie LL, Smith P, Hutchison D, Banfield M, Leopold H, Jabbar A, Sterbenz JPG. From detection to remediation: A self-organized system for addressing flash crowd problems. In: Proc. of the IEEE Int'l Conf. on Communications. Beijing, 2008. 5809-5814.

引证文献1

二级引证文献4

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