基于知识图谱的数据中心异常数据流检测系统设计  

Design of abnormal data flow detection system in data center based on knowledge graph

作　　者：李林 马芳平 彭放 孙延黎 徐镭梦 LI Lin;MA Fangping;PENG Fang;SUN Yanli;XU Leimeng(Guodian Dadu River Drainage Area Hydroelectricity Development Co.,Ltd.,Chengdu 610095,China)

机构地区：[1]国能大渡河流域水电开发有限公司,四川成都610095

出　　处：《电子设计工程》2023年第7期77-81,共5页Electronic Design Engineering

基　　金：国能大渡河流域水电开发有限公司科技创新项目(GJNY-DDH-2020-009)。

摘　　要：为增大主机元件中每秒新建数据流检测链接的数量,在准确分类常规数据与异常数据的同时,按需检测数据中心异常数据流,设计基于知识图谱的数据中心异常数据流检测系统。在IDS主机体系中,分别连接数据辨识模块与流量查询模块,完成系统硬件执行环境的搭建。根据检测信息查询原则,构建完整的知识图谱,设置数据流链接条件,完成系统软件执行环境的搭建,实现基于知识图谱的数据中心异常数据流检测系统的设计与应用。实验结果表明,所提系统主机元件中每秒新建的数据流检测链接数量明显增大,能够精准分类常规数据与异常数据,从而实现对数据中心异常数据流的按需检测。In order to increase the number of new data flow detection links per second in the host components,accurately classify the conventional data and abnormal data,and detect the abnormal data flow in the data center on demand,a data center abnormal data flow detection system based on knowledge map is designed.In the IDS host system,the data identification module and traffic query module are connected respectively to complete the construction of the system hardware execution environment.According to the detection information query principle,build a complete knowledge map,set data flow link conditions,complete the construction of system software execution environment,and realize the design and application of abnormal data flow detection system in data center based on knowledge map.The experimental results show that the number of new data flow detection links per second in the host element of the proposed system increases significantly,which can accurately classify conventional data and abnormal data,so as to realize the on⁃demand detection of abnormal data flow in the data center.

关 键 词：知识图谱 数据中心 数据流检测 IDS主机 流量查询 数据流链接 

分 类 号：TN919[电子电信—通信与信息系统]