基于随机高级Petri网的主动自调度集群系统的性能分析

doi:1000-436X(2006)12-0026-06

通信学报 ›› 2006, Vol. 27 ›› Issue (12): 26-31.doi: 1000-436X(2006)12-0026-06

基于随机高级Petri网的主动自调度集群系统的性能分析

朱巧明^1,²,刘钊^1,²,李培峰^1,²,王汝传^2,³

¹ 苏州大学计算机科学与技术学院，江苏苏州 215006
² 江苏省计算机信息处理新技术重点实验室，江苏苏州 215006
³ 南京邮电大学计算机学院，江苏南京 210003

出版日期:2006-12-25 发布日期:2017-06-22
基金资助:
国家自然科学基金资助项目;江苏省自然科学基金资助项目;江苏省高技术研究计划;江苏省高技术研究计划;江苏省教育厅自然基金资助项目

Performance analysis of ASACS based on SHLPN

Qiao-ming ZHU^1,²,Zhao LIU^1,²,Pei-feng LI^1,²,Ru-chuan WANG^2,³

¹ College of Computer Science and Technology, Suzhou University, Suzhou 215006, China
² Key Lab. of Computer Information Processing Technology of Jiangsu Province, Suzhou 215006,China
³ College of Computer, Nanjing University of Posts and Telecommunications,Nanjing 210003,China

Online:2006-12-25 Published:2017-06-22
Supported by:
The National Natural Science Foundation of China;The Natural Science Foundation of Jiangsu Province;The High-tech Research Project of Jiangsu Province;The High-tech Research Project of Jiangsu Province;The Natural Foundation of Department of Education of Jiangsu Province

摘要/Abstract

摘要：

首先在体系结构上对主动自调度集群系统(ASACS)与传统Web集群服务器系统进行了比较；然后提出了主动自调度集群系统的随机高级Petri网(SHLPN)模型，并设计了模型的精化方案；接着为传统Web服务器集群系统中2种负载均衡调度策略和ASACS的主动自调度策略进行了建模；最后利用SPNP工具对服务器集群系统中的3种调度策略在吞吐量、响应时间及拒绝概率等性能上作了数值分析，发现采用主动自调度策略实现的集群系统能更好地满足QoS的要求。

关键词: 主动自调度集群系统, 随机高级Petri网, 性能分析, 性能模型

Abstract:

After comparing the architecture of automatic self-allocating cluster system (ASACS) with the architecture of traditional system of Web servers cluster, a novel stochastic high-level Petri nets (SHLPN) model for the ASACS was proposed, and the refined design scheme for the model was presented. Also two models of load-balance scheduling poli-cies in the traditional system of Web servers cluster and the model of scheduling policy in ASACS were described. At last, the throughput and the response time and the refuse probability for the three kinds of scheduling policies in server cluster system via the tool of SPNP were analyzed numerically. It was proved that the server cluster system which adopts the automatic self-allocating (ASA) schedule policy can satisfy with the requirement of QoS rather than the traditional sys-tem of Web servers cluster which adopts the load-balance scheduling policy.

Key words: ASACS, SHLPN, performance analysis, performance model

朱巧明,刘钊,李培峰,王汝传. 基于随机高级Petri网的主动自调度集群系统的性能分析[J]. 通信学报, 2006, 27(12): 26-31.

Qiao-ming ZHU,Zhao LIU,Pei-feng LI,Ru-chuan WANG. Performance analysis of ASACS based on SHLPN[J]. Journal on Communications, 2006, 27(12): 26-31.

图/表 9

图1

图2

图3

图4

图5

图6

图7

图8

表1

参考文献 15

[1]	CHOI E . Performance test and analysis for an adaptive load balancing mechanism on distributed server cluster systems[J]. Future Generation Computer Systems, 2004,20(2):237-247.
[2]	KAARIO K , HAMALAINEN T , ZHANG J . Tuning of QoS aware load balancing algorithm (QoS-LB) for highly loaded server clusters[A]. Computer Science LCWS[C]. 2001.250-258.
[3]	ZHANG B Y , MO Z Y , YANG G W . An efficient dynamic load-balancing alborithm in a large-scale cluster[A]. LNCS[C]. 2005.174-183.
[4]	王晓川，叶超群，金士尧．一种基于分布式调度机制的集群体系结构[J]．计算机工程 2002,28(8):232-234. WANG X C , YE C Q , JIN S Y . A new cluster architecture based on distributed schedule mechanism[J]. Computer Engineering, 2002,28(8):232-234.
[5]	王晓川．主动式集群网络服务器调度机制的研究[D]．国防科技大学计算机学院 2001. WANG X C . Research on the Schedule Mechanism of Positive Cluster Internet Server Architecture(PCISA)[D]. National University of Defense Technology, 2001.
[6]	LIN C , MARINESCLL D C . Stochastic high-level Petri nets and applications[J]. IEEE Transactions on Computers, 1988,37(7):815-825.
[7]	单志广，戴琼海，林闯． Web 请求分配和选择的综合方案与性能分析[J]．软件学报， 2001,12(3):355-366. SHAN Z G , DAI Q H , LIN C . In tegrated schemes of Web request dispatching and selecting and their performance analysis[J]. Journal of Software, 2001,12(3):355-366.
[8]	单志广，林闯，魏丫丫．广域Web集群的随机高级Petri网模型及性能分析[J]．系统仿真学报 2003,8(15):93-98. SHAN Z G , LIN C , WEI Y Y . Modeling and performance analysis of globally distributed Web-server clusters using stochastic high-level Petri nets[J]. Journal of System Simulation, 2003,8(15):93-98.
[9]	SHAN Z G , LIN C , MARINESCU D C , et al. Modeling and perform-ance analysis of QoS-aware load balancing of Web server clusters[J]. Computer Networks, Elsevier Science, 2002,40(2):235-256.
[10]	ALLEN A O, Probability, Statistics, and Queueing Theory With Computer Science Applications,2nd ed[M]. New York: Academic, 1990.
[11]	盛友招．排队论及其在计算机通信中的应用[M]．北京：北京邮电大学出版社， 1998.10-102. SHENG Y Z . Queuing Theory and Application of It in Computer Communication[M]. Beijing: Beijing University of Posts and Telecommunications Press, 1998.10-102.
[12]	顾启泰．离散事件系统建模与仿真[M]．北京：清华大学出版社， 1999.14-23. GU Q T . Modeling and Simulation of Discrete Events System[M]. Beijing: Tsinghua University Press, 1999.14-23.
[13]	SUN H , ZANG X . Performance of broadcast and unknown server (BUS) in ATM LAN emulation[J]. IEEE/ACM Transactions on Net-working, 2001,9(3):361-372.
[14]	KISHOR S . SPNP user's manual version 6.0[EB/OL]. , 1999.
[15]	章文嵩．可伸缩网络服务的研究与实现[D]．国防科技大学计算机学院， 2000. ZHANG W S . The Study and Implementation of Scalable Network Services[D]. National University of Defense Technology, 2000.

性能指标	方案
性能指标	FSPF	E-FSPF	ASA
吞吐量(应答数/s)	396.066	396.898	396.094
拒绝概率	0.978％	0.77％	0.953％
响应时间/s	0.194	0.164	0.083

基于随机高级Petri网的主动自调度集群系统的性能分析

Performance analysis of ASACS based on SHLPN

在线阅读

PDF下载

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 9

参考文献 15

相关文章 15

Metrics

推荐阅读 0

[1]	黎赛, 杨亮, 崔琪楣, 于思源. RIS辅助的混合RF/THz系统性能分析[J]. 通信学报, 2022, 43(1): 49-58.
[2]	王文鼐, 张延贺, 吴炜, 柏琛, 王斌. 分布式排队中退避树的深度优先遍历算法[J]. 通信学报, 2021, 42(2): 72-80.
[3]	邓璐,贾焰,方滨兴,周斌,张涛,刘心. 分布式环境下话题发现算法性能分析[J]. 通信学报, 2018, 39(8): 176-184.
[4]	陆凤波,黄知涛,姜文利. 基于Fast-ICA的CDMA信号扩频序列盲估计及性能分析[J]. 通信学报, 2011, 32(8): 136-142.
[5]	鲁蔚锋,吴蒙. 两跳中继CDMA蜂窝系统的上行链路容量分析[J]. 通信学报, 2010, 31(9): 47-54.
[6]	喻莉,白云,朱光喜. 基于LFSN自相似流量模型的随机突发边界分析[J]. 通信学报, 2010, 31(5): 16-21.
[7]	时向泉,苏金树,陈一骄,吴纯青. 网络处理器并行性能模型及多线程停顿特性[J]. 通信学报, 2007, 28(9): 53-59.
[8]	魏安全,沈连丰. 周期序列DSSS系统抗CW干扰性能分析[J]. 通信学报, 2007, 28(5): 1-6.
[9]	刘军,廖建新,朱晓民,王纯. 用于移动计算环境的口令认证协议研究[J]. 通信学报, 2007, 28(5): 36-42.
[10]	盛凌志,皇甫伟,雷文,吴志美. 多类别终端无线局域网实时业务性能研究[J]. 通信学报, 2006, 27(4): 16-22.
[11]	朱国东,宁红宙,刘云,何德全. PKI体系下的网络安全性能评估算法[J]. 通信学报, 2006, 27(3): 119-123.
[12]	张琳,秦家银. Turbo码在无信道状态信息的Nakagami衰落信道中的性能分析[J]. 通信学报, 2006, 27(12): 43-47.
[13]	赵国生,王慧强,李明军. 可生存性网络的形式化建模与分析方法[J]. 通信学报, 2006, 27(11A): 48-52.
[14]	吴乃星,廖建新,徐鹏,朱晓民. 基于软交换的集群媒体服务器的性能评价Petri网模型[J]. 通信学报, 2005, 26(5): 73-79.
[15]	黄少滨,黄辉春,吴艳霞. TCP Vegas与TCP Reno混存网络环境性能分析及改进[J]. 通信学报, 2005, 26(1A): 31-33.