智能网业务控制点负载平衡的研究
详细信息    本馆镜像全文|  推荐本文 |  |   获取CNKI官网全文
摘要
智能网是一个能快速、方便、灵活、经济、有效地生成和实现各种新业务与修改原有业务的体系结构,它是一个业务控制体系结构,是开放式、分布式、与业务无关的通信通用平台。业务控制点和业务交换点是智能网的主要构成部分,它们通过No.7信令进行通信。业务控制点接收业务交换点的请求并进行分析和处理,最后返回给业务交换点请求的响应。
     在智能网中,业务控制点是由多个模块构成的分布式处理系统,用户数据分布在各个模块中,由于用户数据流量的差异导致各个模块间负载的不同。因此,必须平衡各个模块的负载,提高模块的利用率,进而提高整个系统的效率。已有的负载平衡方法——记录迁移方法,它是以记录为单位将用户数据从负载高的模块迁移到负载低的模块来达到平衡模块间负载的效果。但是它不能够动态平衡负载,这种方法主要存在以下两个缺陷:
     1.记录迁移时系统开销问题;
     2.记录迁移循环问题。
     本论文针对上述两个问题,提出一种改进的负载平衡方法,通过对用户数据流量随时间变化的仿真,比较了记录迁移方法改进前后每一模块的流量变化与所有模块平均流量变化的趋近程度,同时还对几个重要参数:模块数量缩减率、负载不平衡比和单位时间内记录迁移数量,它们之间的关系进行讨论,得出改进后的记录迁移方法能够达到动态平衡负载的效果。
The IN (Intelligent Network), which is being investigated as a future communication network, facilitates both rapid introduction of new services and easy modification of existing services. The SCP (Service Control Point) and the SSP (Service Switch Point) are IN components that connected by SS7 (Signaling System No.7). The SCP fields query from the SSP, parse them and perform real-time database query processing, and provides the appropriate responses.
    The SCP allocates data for customers among multiple modules. In the SCP, load can vary among the modules due to dispersion in the amount of traffic for each customer. It is therefore important to balance out this load variation and raise the utilization of each module in order to achieve an efficient distributed processing system. There is a method for balancing the load among modules by transferring customer data in units of records from high-load modules to low-load modules. But the method can not balance load dynamically. There are two limitations in this method:
    1. Processor overhead for transferring record
    2. Circulation for transferring record
    Then, based on the above limitations, this paper proposes a dynamic Load Balancing method. Through the simulation of the change over time of the traffic of customer data, it draws a comparison in the deviation of the change over time of the traffic in a certain SDM (Service Data Module) with respect to that of the average traffic of all modules between old method and new method. The load imbalance ratio, the number of records transferred per unit time and the reduction of number of modules are evaluated. And the relationship among them is also evaluated. Therefore, the dynamic transferring of records is an effective scheme for balancing load among modules in the service control point of the IN.
引文
[1] Yi-Bing Lin, Herman Chung-Hwa Rao, Number Portability for Telecommunication Networks, IEEE Network, Jan. /Feb. 1999, pp. 56-62.
    [2] Carmelita Gorg, Martin Guntermann, and Stephan Kleier, Future Systems for Personal Mobility Services: Design, Performance Evaluation, and Implementation, IEEE Journal on Selected Areas in Communications, Oct. 1997, Vol.15, No. 8, pp. 1672-1683.
    [3] Marcel Mampaey and Alban Couturier, Using TINA Concepts for IN Evolution, IEEE Communications Magazine, Jun. 2000, Vol. 38, No. 6, pp. 94-99.
    [4] Rob Brennan, Brendan Jennings, Conor McArdle, and Thomas Curran, Evolutionary Trends in Intelligent Networks, IEEE Communications Magazine, Jun. 2000, Vol.38, No. 6, pp. 86-93.
    [5] Mark Finkelstein, Jim Garrahan, David Shrader, and Geri Weber, The Future of the Intelligent Network, IEEE Communications Magazine, Jun. 2000, Vol.38, No. 6, pp. 100-106.
    [6] M. Hirano, T.Shiozawa, Y.Kinouchi, and T.Suzuki, Dependency of Load Imbalance on Data Allocation among Distributed Processing Modules in the Intelligent Network, IEICE Transactions, Feb. 1997, Vol. J80-B-I, No. 2, pp. 87-97.
    [7] E. Cancer, R. McCann, and M. Aboudharam, IN Rollout in Europe, IEEE Communications Magazine, Mar. 1993, pp. 38-47.
    [8] Ryoichi Kawahara, Takuya Asaka, and Shuichi Sumita, Overload Control for the Intelligent Network and Its Analysis by Simulation, IEICE Transactions on Communication, Apr. 1995, Vol. E78-B, No. 4, pp. 494-503.
    [9] Ian Dufour, Network Intelligence-The shift ing Balance, Journal of IBTE,
    
    Jan-Feb, 2000, Vol. 1, No. 1, pp83-86.
    [10]Maureen O'Reilly Roche, Call Party Handling Using the Connection View State Approach: A Foundation for Intelligent Control of Multiparty Calls, IEEE Communications Magazine, Jun. 1998, pp. 60-66.
    [11]Fotis G. Chatzipapadopoulos, Menelaos K. Perdikeas, and Iakovos S. Venieris, Mobile Agent and CORBA Technologies in the Broadband Intelligent Network, IEEE Communications Magazine, Jun. 2000, pp. 116-123.
    [12]George N. Prezerakos, Stefano Salsano, Alexander W. van der Vekens, Fabrizio Zizza, INSIGNIA: A Pan-European Trial for the Intelligent Broadband Network Architecture, IEEE Communications Magazine, Jun. 1998, pp. 68-76.
    [13]张孟顺,智能网的网间互通,世界网络与多媒体,1999,第七卷,第十一期,22-25。
    [14]杨孟辉,廖建新,陈俊亮,中国智能网系统及其智能业务,世界网络与多媒体,1999,第七卷,第十一期,10-15。
    [15]卢美莲,七号信令网的应用及发展前景,世界网络与多媒体,1998,第六卷,第十一期,8-13。
    [16]郭春源,我国智能网现状与发展,世界网络与多媒体,2000,第八卷,第二期,14-18。
    [17]刘多,ITU-T对智能网的研究(上),世界网络与多媒体,2000,第八卷,第2期,29-32。
    [18]刘多,ITU-T对智能网的研究(下),世界网络与多媒体,2000,第八卷,第3期,28-34。
    [19]’97年度,“863”通信新技术跟踪研究报告之二,智能网技术,1998,第七期,13-15。
    [20]刘云生,李国徽,实时内存数据库的装入,软件学报,2000,第十一期,第六卷,829-835。
    [21]王于同,李剑,并行数据库负载平衡算法研究,计算机工程与应用,1998,第十一期,6-8。
    
    
    [22]刘欢,分布式处理系统中进程调度的设计,计算机工程与应用,1999,第八期。
    [23]孙晓蓉,徐春光,王育民,网络和分布式系统中的认证,计算机研究与发展,1998,第十期,第三十五卷。
    [24]龚志宏,电信网络的演进,世界网络与多媒体,2000,第八卷,第五期,35-39。
    [25]盛骤,谢式千,概率论与数理统计,第2版,1989,高等教育出版社。
    [26]龚双瑾,智能网技术,第1版,1999,人民邮电出版社。
    [27]杨放春等,智能化现代通信网,第1版,1999,北京邮电大学出版社。
    [28]李晓峰、勾学荣,智能网,第1版,1996,人民邮电出版社。
    [29]李晓峰,智能网技术,第1版,1998,北京邮电大学出版社。
    [30]纪红,7号信令系统,第2版,1999,人民邮电出版社。
    [31]郑振楣,于戈,郭敏,分布式数据库,第1版,1998,科学出版社。
    [32]张志涌,刘瑞桢,杨祖樱,掌握和精通MATLAB,第1版,1997,北京航空航天大学出版社。
    [33]逯昭义,王思明,计算机通信网信息量理论,第1版,1997,电子工业出版社。
    [34]黄锡伟,朱秀昌,宽带通信网络,第1版,1998,人民邮电出版社。
    [35]Andrew S.Tanenbaum,计算机网络,熊桂喜,王小虎,第3版,1998,清华大学出版社。
    [36]王珊,陈红,数据库系统原理教程,第1版,1998,清华大学出版社。
    [37]Jean Walrand,Pravin Varaiya,高性能通信网络,韩毅刚,严春莹,吴英,第1版,2000,电子工业出版社。
    [38]周龙骥,分布式数据库管理系统实现技术,第1版,1998,科学出版社。
    [39]David M.Kroenke,数据库处理:基础、设计与实现,施泊乐,顾宁,雍信阳,范志新,王轶,第1版,1998,电子工业出版社。

© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号

地址:北京市海淀区学院路29号 邮编:100083

电话:办公室:(+86 10)66554848;文献借阅、咨询服务、科技查新:66554700