【作者】 孙黎丽； 陈雪； 刘冬；

【Author】 Sun Lili~1 Chen Xue~2 Liu Dong~3 1,2.Key Laboratory of Optical Communication and Lightwave Technologies(Beijing University of Posts and Telecommunications),Ministry of Education,Beijing 100876,China 3.Beijing GW Technologies Co.,Ltd.,Beijing 100085,China

【机构】 北京邮电大学光通信与光波技术教育部重点实验室； 北京格林威尔科技发展有限公司；

【摘要】 在网络发生故障后,为了尽快排除故障并恢复业务,需要快速定位故障原因。而如何对故障产生的大量告警数据进行过滤、分析并生成有效的故障定位提示信息,提高网络的稳定性和可靠性,则日益成为网络运营商关注的重要问题。结合了数据库技术和人工智能技术优点的知识发现技术(KDD：Knowledge Discovery in Databases)的出现正好可以适应这一需求,它能有针对性地快速定位故障,缩短业务中断时间。 KDD 通过剔除不必要和不相关的信息,实现对大量的告警进行合并和转化,使呈现给网管操作人员的告警信息数量大大减少,从而提高了信息的有效性,帮助网管人员找出产生告警的真正原因,进而快速定位故障并恢复服务。KDD 甚至还能预测故障的发生从而指导维护人员提前介入,防止故障真正发生,实现主动运维。因此,它可以大大提高故障管理的水平。更多还原

【Abstract】 In order to recover the failures in network,fast fault location is necessary in network management for the sake of decreasing the time of service interrupt.So the method of extracting useful information from blocks of alarm data has played more and more important role in network management to enhance the stability and reliability of the network.A new knowledge discovery technology called KDD(Knowledge Discovery in Databases)emerges to solve the problem,which takes advantage of both data base technology and artificial intelligence technology.By removing unnecessary and irrelevant data information,KDD can convert the blocks of alarm data to a small amount of key information which is then submitted to the network management platform and network operators.Therefore, it can realize fast fault location for service recovery by means of increasing the efficiency of information. Furthermore,KDD can perform assistance to predict the possible emergence of failure and realize the pre-protection by analyzing the key information,consequently it can enhance the stability and reliability of the whole network.更多还原