分布式应用系统日志模型解析  

ANALYSIS OF DISTRIBUTED APPLICATION SYSTEM LOG MODEL

作　　者：李贝贝[1] Li Beibei(China Academy of Railway Sciences,Beijing 100081,China)

机构地区：[1]中国铁道科学研究院,北京100081

出　　处：《计算机应用与软件》2023年第11期24-28,共5页Computer Applications and Software

基　　金：中国国家铁路集团有限公司科技研究开发计划课题(N2019X016);中国铁道科学研究院集团有限公司科研项目(2019YJ122)。

摘　　要：为了使企业从已有的信息化分布式应用实例中沉淀公共的日志处理方法,实现日志的快速管理与分析,提出一套日志从生产到消费的全流程解析方法。梳理应用业务并分解出日志种类;确定不同种类日志中的行字段内容与字段类型,在此基础上设计日志统一模型,将日志内容分别写到不同的日志文件;结合开源组件,建立日志的分布式监测、过滤与存储分析系统,实现日志的统一管理与准实时挖掘分析。通过铁路人脸识别系统的日志应用表明,该模型具有很好的可读性、稳定性和扩展性,能快速实现分布式应用日志的统一管理与分析。To make enterprises accumulate common log processing methods from existing distributed application examples of information,as well as realize the rapid management and analysis of logs,this paper proposes a set of log analysis method from production to consumption.It sorted out application services and log categories.It determined the content and type of texts in different logs to design the unified model for log on this basis.It wrote log contents to different log files,established the distributed monitoring,filtering and storage analysis system of log,which realized the unified management and quasi real-time mining analysis of logs.According to the application of the log in the railway face recognition system,it is found that the model has good readability,stability and expansibility,which can quickly realize the unified management and analysis of distributed application logs.

关 键 词：分布式应用 日志模型 日志生产 日志消费 铁路人脸识别 

分 类 号：TP311.5[自动化与计算机技术—计算机软件与理论]