云虚拟机异常检测场景下改进的LOF算法  

Improved LOFAlgorithm in Cloud Virtual Machine Anomaly Detection Scenario

作　　者：贺寰烨 林果园[1,2,3] 顾浩 方梦华 HE Huanye;LIN Guoyuan;GU Hao;FANG Menghua(School of Computer Science and Technology,China University of Mining and Technology,Xuzhou,Jiangsu 221116,China;Mine Digitization Engineering Research Center of the Ministry of Education,Xuzhou,Jiangsu 221116,China;State Key Laboratory for Novel Software Technology,Nanjing University,Nanjing 210023,China)

机构地区：[1]中国矿业大学计算机科学与技术学院,江苏徐州221116 [2]矿山数字化教务部工程研究中心,江苏徐州221116 [3]南京大学计算机软件新技术国家重点实验室,南京210023

出　　处：《计算机工程与应用》2020年第23期80-86,共7页Computer Engineering and Applications

基　　金：软件新技术国家重点实验室开放基金(No.KFKT2018B27);中央高校基础研究基金(No.2017XKQY079)。

摘　　要：针对云服务中由于资源超额预定造成负载不均衡的云虚拟机异常,提出了一种基于密度空间的局部离群因子(Local Outlier Factor Based on Density Space,LOFBDS)算法。LOFBDS算法参考DBSCAN(Density-Based Spatial Clustering of Applications with Noise)算法,将云虚拟机在密度空间中的性质融合至LOF算法之中,提出对云虚拟机的判断规则,以达到优化对正常云虚拟机的检测过程,提高检测效率。实验结果表明,所提出的算法对云服务负载不均造成的云虚拟机异常有着良好的检测效率,并且时间花费较少。In this paper,a LOFBDS(Local Outlier Factor Based on Density Space)algorithm is proposed to detect the cloud virtual machine anomaly with unbalanced load caused by resource overbooking in cloud services.The LOFBDS algorithm refers to the DBSCAN(Density Based Spatial Clustering of Applications with Noise)algorithm.It integrates the properties of cloud virtual machines in the density space into the LOF algorithm,and proposes judgment rules for cloud virtual machines to optimize the normal cloud virtual machine to improve detection efficiency.Experimental results show that the algorithm proposed in this paper has a good detection efficiency and less time cost for cloud virtual machine exceptions caused by uneven cloud service load.

关 键 词：云服务 云虚拟机 异常检测 局部离群点检测 

分 类 号：TP306.3[自动化与计算机技术—计算机系统结构]