智能化网格电信系统的故障预测方法

doi:10.11959/j.issn.1000-0801.2018118

摘要/Abstract

摘要：

尝试用基于深度学习的相关人工智能技术，分析服务器集群上的进程和端口网络，并对网络节点进行状态预测。具体地，结合运维过程中的先验知识对网络节点的特征进行细致选择，预测网络中各个进程和端口的异常（崩溃）状态。实验结果表明，进程节点的运行信息（如 CPU 和内存使用率）、进程间的通信情况以及进程节点在整个网络中的结构特征对于判断该节点的状态具有一定的指导价值，而这些特征在时间维度上的变化量同样反映了进程/端口的状态。

关键词: 故障预测, 深度学习, 二分类

Abstract:

Some approaches based on deep learning would be used to analyze the process and port network on a server cluster.Specifically,the features of nodes were carefully selected in server cluster network,by combining the prior knowledge from actual operations,and the abnormal state of processes or ports on the cluster was predicted.According to the research,the running information such as loads of CPU and memory,communications between processes and the structural features in the process network was valuable in predicting the states of processes and ports; furthermore,the changes of features mentioned above in the time dimension reflected the states of processes or ports,too.

Key words: fault prediction, deep learning, binary classification

中图分类号:

TP391.1

蔡珩,戈磊. 智能化网格电信系统的故障预测方法[J]. 电信科学, 2018, 34(6): 183-191.

Heng CAI,Lei GE. Intelligent fault prediction method of telecom system[J]. Telecommunications Science, 2018, 34(6): 183-191.

图/表 11

图1

图2

图3

图4

图5

图6

图7

图8

图9

图10

表1

参考文献 23

[1]	DAMANI O P , CHUNG P E , HUANG Y ,et al. ONE-IP:techniques for hosting a service on a cluster of machines[J]. Computer Networks and ISDN Systems, 1997,29(8-13): 1019-1027.
[2]	MOOLENBROEK D V C , APPUSWAMY R , TANENBAUMA S . Integrated system and process crash recovery in the loris storage stack[C]// Networking,Architecture and Storage(NAS),2012 IEEE 7th International Conference,October 14-17,2012,Seoul,Korea (South). Piscataway:IEEE Press, 2012: 1-10.
[3]	MOHAMED E , ABDEL-WAHAB H , SALAMA I . Multicast address management in the internet:a study of the port blocking problem[R]. 1999.
[4]	NEWMAN M , BARABASI A L , WATTS D J . The structure and dynamics of networks[M]. Princeton: Princeton University PressPress, 2006: 419-421.
[5]	GRANOVETTER M . The strength of weak ties[J]. American Journal of Sociology, 1973,78(6): 1360-1380.
[6]	ONNELA J P , SARAMAKI J , HYVONEN J ,et al. Structure and tie strengths in mobile communication networks[J]. The National Academy of Sciences, 2007,104(18): 7332-7336.
[7]	CHORMANSKI K , MATUSZAK M , MIEKISZ J . Scale-free graph with preferential attachment and evolving internal vertex structure[J]. Journal of Statistical Physics, 2013,151(6): 1175-1183.
[8]	MILGRAM S . The small world problem[J]. Psychology Today, 1967,2(1): 185-195.
[9]	FREEMAN L . A set of measures of centrality based on betweenness[J]. Sociometry, 1977,40(1): 35-41.
[10]	NEWMAN M . Networks:an introduction[M]. Oxford: Oxford University PressPress, 2010.
[11]	GOMEZ-RODRIGUEZ M , LESKOVEC J , KRAUSE A . Inferring networks of diffusion and influence[J]. ACM Transactions on Knowledge Discovery from Data, 2010,5(4): 1-37.
[12]	RODRIGUEZ M G , BALDUZZI D , SCHOLKOPF B . Uncovering the temporal dynamics of diffusion networks[C]// The 28th International Conference on Machine Learning (ICML),June 28-July 2,2011,Bellevue,Washington,USA.[S.l.:s.n. ], 2011: 561-568.
[13]	LIBEN-NOWELL D , KLEINBERG J . The link prediction problem for social networks[J]. Journal of the Association for Information Science ＆ Technology, 2007,58(7): 1019-1031.
[14]	LAST M . Kernel methods for pattern analysis[M]. Beijing: China Machine PressPress, 2005.
[15]	KIM I . Convolutional neural networks for sentence classification[J]. arXiv:1408.5882, 2014
[16]	CONKLIN J D . Applied logistic regression[J]. Technometrics, 2013,44(1): 81-82.
[17]	BRANDES U . A faster algorithm for betweenness centrality[J]. Journal of Mathematical Sociology, 2001,25(2): 163-177.
[18]	SABIDUSSI G . The centrality index of a graph[J]. Psychometrika, 1966,31(4): 581-603.
[19]	JAPKOWICZ N , . The class imbalance problem:significance and strategies[C]// The 2000 International Conference on Artificial Intelligence(IC-AI’2000),June 26-29,2000,Las Vegas,USA.[S.l.:s.n]. 2000.
[20]	GEISSER S . Predictive inference:an introduction[M]. New York: Chapman and HallPress, 1993.
[21]	KOHAVI R , . A study of cross-validation and bootstrap for accuracy estimation and model selection[C]// The Fourteenth International Joint Conference on Artificial Intelligence,August 20-25,1995,Montreal,Quebee,Canada. New York:ACM Press, 1995: 1137-1143.
[22]	LEIKE A .Demonstration of the exponential decay law using beer froth[J]. European Journal of Physics, 2002,23(1): 2-1.
[23]	MOHAR B . A linear time algorithm for embedding graphs in an arbitrary surface[J]. SIAM Journal on Discrete Mathematics, 2006,12(1): 6-26.

测试日期	正负样本比	准确率	召回率	F1得分
4月14日	0.850%	0.941	0.640	0.762
4月15日	0.882%	0.973	0.655	0.783
4月16日	0.751%	0.953	0.891	0.921
4月17日	0.669%	0.971	0.805	0.880
4月18日	0.703%	0.723	0.791	0.756
4月19日	1.42%	0.927	0.432	0.589
4月20日	0.866%	0.951	0.735	0.830