摘要
随着万维网的迅猛发展,宽带的快速普及,使得网民数量激增,各种基于互联网的应用层出不穷。这一切要求网络服务器提供更高的数据吞吐量并向用户提供更低的用户可察觉延迟并最终导致集群技术的广泛使用。集中式集群比分布式集群具有更高负载均衡效果,但是集中式集群的一个严重的不足是前端均衡器易成为系统性能瓶颈。由于均衡器是系统对外的唯一接入点,它不仅承担着对集群后端服务器群的负载监测任务,同时又必须完成负载分配任务。因此,当网络流量很大时,它就容易成为系统的瓶颈,严重时甚至造成系统的崩溃。
为了解决这一问题,本文主要完成以下的工作:
(1)本文针对基本的负载均衡算法加权轮转算法所存在的不足,提出一种基于离散型均匀分布的原理的基本负载均衡调度策略。它通过后端服务器向均衡器发送不同速率的请求流来承载各负载节点的负载状况来降低调度算法的复杂度,使得前端均衡器的负载监测任务后移到后端各服务器,从而降低了前端均衡器的工作压力。
(2)对所提出负载均衡策略引入动态调度机制,不仅拓宽了基于离散性均匀分布的负载均衡策略的应用范围,而且进一步提高了负载均衡的运行效率。
(3)根据计算机系统的体系结构特点,设计了一种测试服务器动态负载的方法。通过对影响服务器负载变化的基本因素的分析,给出评估服务器综合负载,动态负载的算法。
(4)运用网络仿真工具OPNET,搭建集群系统模型,设计负载均衡模块,并对上述工作内容进行仿真测试。结果表明基于离散型均匀分布的负载均衡策略具有较好的负载均衡效果,能有效地降低前端均衡器的工作负载。
本文成果可用于集中式集群中,用来减轻前端均衡器的工作压力,降低其成为系统瓶颈的可能性。
The rapid growth of the World Wide Web, the fast popularization of Broadband, the increasing quantity of netizen, and various web-based applications are creating demand for high performance web servers to offer better throughput and shorter user-perceived latency. This demand leads to widely used cluster-based web servers in the Internet infrastructure. Centralized cluster is more effective at load balancing than distributed cluster, but centralized cluster contains a serious weakness, system bottlenecking. This is because a front-end dispatcher existing in the cluster system is the only access point, which not only bears monitoring the load states from the back-end servers, but also must complete distributing the load among those servers. Therefore, when network traffic is large, it will easily become the bottleneck in the system, and even cause the collapse of the system in severe cases.
In order to solve this problem, these tasks in the following are completed in this paper.
First, a basic scheduling strategy of load-balancing based on principle of a discrete uniform distribution is proposed to solve the deficiencies of Weighted Round Robin load balancing algorithm in this paper. In this scheduling, each server sends the ask stream at different rates to the balancer and these streams can carry the load conditions, so the balance does not monitor the workload while the back-end servers do, and as a result is that the strategy achieves reducing the work pressure on balancer.
Second, dynamic scheduling mechanism is introduced and used in the proposed strategy on load balancing, and it does not only broaden the scope of application of load balancing strategy, but also improve the operating efficiency of the load balancing.
Third, A method for sever is presented to calculate dynamic load of server based on the characteristics of Computer system architecture, which evaluates a comprehensive load and dynamic load of server by analyzing the basic elements of affecting the sever load.
Finally, by the network simulation tool, OPNET, this paper builds cluster models, designs load balancing module, and examines the content of above work. The results shows that the load-balancing strategy proposed based on the discrete uniform distribution has a good load balancing effect, and it can effectively reduce the workload of front-end balancer.
Research findings in the paper are used in centralized cluster to relieve the work pressure of balancer and the possibility of its becoming the system bottleneck.
引文
[1] Ahmad Faoura, Nashat Mansourb Weblins: A scalable WWW cluster.based server Advances in Engineering Software [M].2006
[2]林闯,单志广,任丰原.计算机网络服务质量(QoS) [M].清华大学出版社,2006.04
[3]熊智,晏蒲柳,郭成城.Web集群服务器中分发静态请求的调度算法.计算机应与研究.2006,12:84-88
[4] http://www.ibm.com/developerworks/cn/linux/cluster/lvs/part1/index.html[EB/OL]2009.9
[5]王琦,孟昭勇.循环复用DNS实现多服务器的负载均衡.计算机与信息技术[J].2006,94:104-105
[6] Valeria Cardellini, Michele Colajanni and Philip S. Yu DNS dispatching algorithms with state estimators for scalable Web.server clusters ACM Portal .World Wide Web[M].1999
[7] Buyya R.高性能集群计算:编程和应用[M]. (郑纬明译).北京:电子工业出版社,2001
[8]郭成城,晏蒲柳.一种异构Web服务器集群动态负载均衡算法计算机学报[J].2005,28(2):179-184
[9]陈志刚,许伟,曾志文.一种基于预测的动态负载均衡模型及算法研究.计算机工程[J]2004,30(23):87-89
[10]王霜,修保新,肖卫东.Web服务器集群的负载均衡算法研究.计算机工程与应用[J].2006.25:78-80,99
[11]李清玉,韦卫,侯紫峰.防火墙集群动态自适应负载均衡算法设计与实现.计算机工程.2007.31(1):56-59
[12]余庆春,周怀北,高山.计算机多用户系统负载测试软件的研究与设计.计算机应用研究[J].2005,3:72-73,75
[13]王田,杨士中.Web服务器的负荷状态检测技术.小型微型计算机系统[J].2005,26(3)
[14]李飞,何炎祥,张晶晶.基于性能预测和Agent的网格负载平衡的实现.计算机工程[J].2005,31(15)104-106
[15] scalable load balancing on distributed Web serves using mobile agents《parallel distribute comput》[M]2003
[16] Keren A, Barak A. Adaptive placement of parallel java agents in a scalable computing cluster[C]. In: proc of the Workshop on Java for High Performance Network Computing, ACM Press, Stanford University. Plao Alto, CA, USA, 1998
[17]程伟,卢泽新,王宏.一种新的服务器集群系统负载均衡技术[J].计算机工程与科学,2006, 28(2):31-32,34
[18]崔小燕.Linux集群系统分析.西安邮电学院学报[J].2006,11(5):103-106
[19]杨磊,郭庆平.负载均衡技术分析及LVS实现武汉理工大学学报(交通科学与工程版)2006.28(1):24-29
[20]张志友.计算机集群技术概述.实验室研究与探索[J].2006,25(5):607-609
[21]杨建军,刘雄.基于Unix的负载均衡集群方案设计[J].计算机工程与设计, 2005,26(7):1897-1899
[22] Dahlia M.Interpreting Stale Load Information[A].Proc of the 19th Int’l Conf On Distributed Computing Systems[c].1999
[23]郭晓明,蔡滨,付康.用动态反馈负载均衡算法计算Web服务器的最大负载量[J].计算机与现代化.2006,11:36-39
[24] http://vip.calis.edu.cn/page8.htm负载均衡技术[EB/OL]2008.12
[25]朱利,张兴军.Web服务器组的负载均衡方法研究.小型微型计算机系统.2003,24(12)
[26]王志晓,牛强.基于双层分配器的Web服务集群负载平衡解决方案.计算机应用[J].2006,26(2) :307-309
[27] Eunmi Choi.Performance test and analysis for an adaptive load balancing mechanism on distributed server cluster systems Future Generation Computer Systems 20 (2004) 237–247
[28]路明怀龚正虎.Linux服务器下多网卡负载均衡的研究与实现计算机与信息技术[J].2006,3:24-26
[29] Cardellini V,Colajanni M. Dynamic load balancing on Web.sever systems[J].IEEE Computing, 1999,3 (3)
[30]任彦琦,等.一种基于内容的web服务器负载均衡算法[J].计算机工程,2005.31(2)
[31] Bryhni H, Klovning E, Kure O. A comparison of load balancing techniques for scalable Web servers [J ].IEEE N etwork, 2000, 14 (4) : 58-64.
[32] VALERIA CARDELLINI,MICHELE COLAJANNI,PHILIP S. YU. DYNAMIC LOAD BALANCINGON Web.SERVER SYSTEMS.IEEE INTERNET COMPUTING.1999:28-39
[33]胡子昂,王立.算法、网络拓扑及调度频率与动态负载平衡的关系[J].计算机工程与科学, 2000, 22(1): 104-107..
[34]王霜,修保新,肖卫东.Web服务器集群的负载均衡算法研究.计算机工程与应用[J].2004.25:78-80
[35]李文正,郭巧,王利,郭为民.Internet服务器负载均衡的研究与实现.计算机工程[J].2005,31(6):97-99
[36] Der.Chiang Lia, Chihsen Wua, Fengming M. Changb.Determination of the parameters in the dynamic weighted Round.Robin method for network loadbalancing.Computers & Operations Research.2005(32), 2129–2145
[37] MinHwan Ok1, Myong.soon Park2.Request Distribution for Fairness with a New Load.Update Mechanism in Web Server Cluster IFIP International Federationfor Information Processing [M],2004
[38]卢琰琰,吴海燕,陈怀楚,高国柱,王映雪.基于联合压力测试的Web应用程序性能预测方法.计算机应用[J].2006,26(6):1472-1474
[39]周佳祥,郑纬民,杨广文.一种基于进程迁移的自适应双阈值动态负载平衡系统.清华大学学报(自然科学版) [J].2000 ,40(3)
[40]彭德巍,何炎祥.基于Agent的负载均衡框架应用研究.计算机工程与应用[J].2005,5:153-156
[41] Mattner, L. and Roos, B. Maximal probabilities of convolution powers of discrete uniform distributions http://arXiv.org/abs/math[EB/OL].(2009-03-12).
[42]杨义先,等.网络信息安全与保密[M].北京:邮电大学出版社,2001
[43]石怀伟,王华,张念军,赵林胜,佀同光.OPNET网络仿真技术及其应用分析.计算机工程与设计[J].2006,27(17):3309-3311.
[44]王文博.张金文.OPNET Modeler与网络仿真[M].人民邮电出版社.2003.10:68-138
[45]刘珩,安建平,杨杰.组播协议在OPNET中的建模与仿真[J].计算机仿
真.2005,22(5):141-144.
[46]张剑.基于OP NE T仿真建模方法研究[ D].武汉:武汉理工大学,2005
[47]陈敏. OPNET网络仿真.清华大学出版社[M] .2004.4:187-220