期刊文献+

集群系统中实现软实时服务机制的研究与评估 被引量:2

Scheduling scheme and performance assessment for timely responsive service on cluster server
下载PDF
导出
摘要 研究了集群架构服务器的三种实时调度策略,即集中式策略、分布式策略和主动型全分布策略。与调度功能主要集中在前端机的集中式策略比较,分布式策略将调度功能部分离散至后端节点,改善了系统的可扩展性。而主动型全分布式调度则进一步实现了调度分派功能的完全离散化,通过后端机的主动拉取事件和频率调控实现了负载平衡的自调节,极大地改善了系统的扩展性和服务效率。通过试验研究和试验表明,这三种都有效地降低了集群的响应服务请求的时间,还保证了系统的输出最大化。 This paper presents three real-time scheduling schemes for cluster servers:centralized scheme,distributed scheme and pull-based distributed scheme.Distributed scheme deploys scheduling on the end-node,while the scheduling work of centralized scheme is on the front-nodes,and that makes the former has a greater scalability.Pull-based distributed scheme makes end-nodes ask for events and direct load balance by themselves,to improve the scalability and efficiency of real-time cluster system.Lots of experimental results have been presented,illustrating the three scheduling schemes are benefit for decreasing the response time and maximizing the system throughout.
出处 《计算机工程与应用》 CSCD 北大核心 2007年第25期133-139,146,共8页 Computer Engineering and Applications
基金 国家高技术研究发展计划(863)( the National High- Tech Research and Development Plan of China under Grant No.2003AA116010) 国家教育部新世纪人才支持计划 (the New Century Excellent Talent Foundation from MOE of China under Grant No.NCET-04-0843) 重庆市自然科学基金(the Natural Science Foundation of Chongqing City of China under Grant No.2005BB2192) 。
关键词 集群服务器 实时服务 调度策略 实时响应特性 cluster server real-time service scheduling schemes real-time response characteristic
  • 相关文献

参考文献16

  • 1Buyya R.High performance cluster computing:architectures and systems:volume I[M].New Jersey:Prentice Hall PTR,1999.
  • 2Aron M,Druschel P,Zwaenepoel W.Cluster reserves:a mechanism for resource management in cluster-based network servers[C]//Proc of 2000 ACM SIGMETRICS Intl Conf on Measurement and Modeling of ComputerSystems,Santa Clara,CA,2000-06.
  • 3Pradham P,Chiueh T.Implementation and evaluation of a QoS-capable cluster-based router[C]//Proc of Super Computing 2000,Dallas,2000-11.
  • 4Bianchini R,Carrera E V.Analytical and experimental evaluation of cluster-based network serves[J].World Wide Web Journal,2000,3(4).
  • 5Martinez K,Hart J K,Ong R.Environmental sensor networks[J].IEEE Computer,August 2004,37(8):50-56.
  • 6Liu C,Layland J.Scheduling algorithms for multiprogramming in a hard real time environment[J].Journal of ACM,1973,20(1):46-61.
  • 7Zhang Y,Sivasubramaniam A.Scheduling best-effort and real-time pipelined applications on time-shared clusters[C]//Proc the 13th ACM Symposium on Parallel Algorithms and Architectures(SPAA'2001),July 2001:209-218.
  • 8Essafi L,Bolch G,de Meer H.Dynamic priority scheduling for proportional delay differentiated services[EB/OL].[2006-05].http://www4.informatik.uni-erlangen.de/TR/ps/TR-I4-01-03.ps.gz.
  • 9Zhang W.Linux virtual server for scalable network services[C/OL]//Ottawa Linux Symposium,2000.http://www.linuxvirtualserver.org/Documents.html.
  • 10Egevang K,Francis P.The IP network address translator(NAT)RFC 1631[S],1994-05.

同被引文献12

引证文献2

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