基于分布式计算的异常网站筛查系统的设计与实现  

Design and implementation of distributed computing based screening system of abnormal websites

作　　者：姜鹏[1] 赵正利[2] 

机构地区：[1]中国海洋大学网络与信息中心,山东青岛266100 [2]中国海洋大学教育系,山东青岛266100

出　　处：《山东科学》2016年第4期106-111,共6页Shandong Science

摘　　要：设计了一种在大型园区环境下对互联网异常网站进行筛查的系统,系统在.NET框架下开发运行,基于分布式计算的模式部署,采用HDFS和SQL Server存储数据。分布在多个子区域的探针根据多设备日志的排序数据进行异常网站的筛查,核心服务部分进行故障汇总分析、任务分发推送和知识库归档等操作。该系统具有一定的故障分析能力,可以根据已有知识库形成故障分析指导,通过调用预定义脚本启用备用设备或线路。系统运行结果表明,设计科学有效,能够给大型企事业单位的网络运维和管理提供参考依据。We devise an automatic screening system for abnormal websites on the Internet in a large-scale zone. It is implemented with . NET Framework, distributed deployment, HDFS and SQL Server. Multiple subareas deployed probes can automatically screen out abnormal sites based on the sorted log data from multiple devices. Its core services include metaanalysis of fault, task distribution, and knowledge base archive. It has certain fault analysis capability, fault analysis and guidance based on existing knowledge b a s e, alternate devices or lines startup by calling predefined scripts. Its operation results show that it is scientific and effective, and can provide a reference for network operation and maintenance of largescale enterprises.

关 键 词：分布式计算 信息系统 B/S模式 .NET 

分 类 号：TP311.1[自动化与计算机技术—计算机软件与理论]