一种基于增量贝叶斯疑似度的事件驱动故障定位算法

张成; 廖建新; 朱晓民

doi:10.3724/SP.J.1146.2008.00610

一种基于增量贝叶斯疑似度的事件驱动故障定位算法

doi: 10.3724/SP.J.1146.2008.00610

基金项目:

国家杰出青年科学基金(60525110)，国家973 计划项目(2007CB307100，2007CB307103)，新世纪优秀人才支持计划(NCET -04-0111)和电子信息产业发展基金项目(基于3G的移动业务应用系统)资助课题

计量
- 文章访问数: 3615
- HTML全文浏览量: 85
- PDF下载量: 851
- 被引次数: 0
出版历程
- 收稿日期: 2008-05-15
- 修回日期: 2008-09-18
- 刊出日期: 2009-06-19

An Event-Driven Fault Localization Algorithm Based on Incremental Bayesian Suspected Degree

摘要

摘要: 现有的故障定位算法大多基于时间窗口，窗口大小设置的合理与否会对算法准确度产生重要影响。为了避免因窗口设置不当造成算法性能的下降，该文以概率加权的二分图作为故障传播模型，提出了一种基于增量贝叶斯疑似度(Incremental Bayesian Suspected Degree，IBSD)的启发式故障定位算法。IBSD算法采用事件驱动的方式依次分析观察到的征兆，通过增量计算对应故障的贝叶斯疑似度，确定当前征兆前提下最有可能的故障集。仿真实验表明，IBSD算法具有较高的故障检测率和较低的故障误检率，即使在部分告警无法观察的情况下，算法依然具有较高的故障检测率。算法具有多项式计算复杂度，可以满足大规模通信网故障定位的要求。
- 故障管理; 故障诊断; 故障定位; 故障传播模型
Abstract: Most fault localization techniques is based on time windows. The size of time windows impacts on the accuracy of the algorithms greatly. This paper takes weighted bipartite graph as fault propagation model and proposes a heuristic fault localization algorithm based on Incremental Bayesian Suspected Degree (IBSD) to eliminate the above shortcomings. IBSD sequentially analyzes the incoming symptoms in an event-driven way and incrementally computes the Bayesian Suspected Degree and determine the most probable fault set for the current observed symptoms. Simulation results show that the algorithm has high fault detection ratio as well as low false positive ratio and has a good performance even in the presence of unobserved alarms. The algorithm which has a polynomial computational complexity could be applied to large scale communication network.

HTML全文

参考文献(1)

Steinder M and Sethi A S. The present and future of eventcorrelation: A need for end-to-end service fault localization[C]. Proc. World Multi-Conf. Systemic, Cybernetics, andInformatics (SCI), Orlando, FL, 2001: 124-129.[2]Steinder M and Sethi A S. A survey of fault localizationtechniques in computer networks [J]. Science of ComputerProgramming, 2004, 53(2): 165-194.[3]Mas C and Thiran P. A review on fault location methods andtheir application to optical networks [J]. Optical NetworksMagazine, 2001, 2(4): 73-87.[4]Mas C and Thiran P. An efficient algorithm for locating softand hard failures in WDM networks [J].IEEE Journal onSelected Areas in Communications.2000, 18(10):1900-1911[5]Zhao Y, Chen Y, and Bindel D. Towards unbiased end-to-endnetwork diagnosis [C]. Proceedings of the ACM SIGCOMM2006 Conference on Applications, Technologies, Architectures,and Protocols for Computer Communications, Pisa, Italy,2006: 219-230.[6]Katzela I and Schwartz M. Schemes for fault identification incommunication networks [J]. IEEE/ACM Trans. onNetworking, 1995, 3(6): 733-764.[7]Yemini S and Kliger S. A coding approach to eventcorrelation, integrated network management [C]. Proceedingsof the Fourth International Symposium on IntegratedNetwork Management, Santa Barbara, California, USA, 1995:266-277.[8]Steinder M and Sethi A S. End-to-end service failurediagnosis using belief networks [C]. Proc. NetworkOperations and Management Symposium (NOMS), Florence,Italy, 2002: 375-390.[9]Steinder M and Sethi A S. Probabilistic fault localization incommunication systems using belief networks [J].IEEE/ACM Trans. on Networking.2004, 12(5):809-822[10]Huang Xiao-hui, Zou Shi-hong, Wang Wen-dong, and ChengShi-duan. Fault management for Internet service: Modelingand algorithms [C]. IEEE International Conference onCommunications, Istanbul, Turkey, June 2006, 2: 854-859.[11]黄晓慧, 邹仕洪, 王文东, 程时端. Internet服务故障管理: 分层模型和算法 [J]. 软件学报, 2007, 18(10): 2584-2594.Huang Xiao-hui, Zou Shi-hong, Wang Wen-dong, and ChengShi-duan. Internet services fault management: Layeringmodel and algorithm. Journal of Software, 2007, 18(10):2584-2594.[12]Steinder M and Sethi A S. Probabilistic event-driven faultdiagnosis through incremental hypothesis updating [C].IFIP/IEEE Eighth International Symposium on IntegratedNetwork Management, Colorado, USA, 2003: 635-648.

施引文献

资源附件(0)

访问统计