The Design And Implementation Of Condition Monitoring System Oriented To Large Scale Distributed Application

With the constant development of Information Technology by Government andvarious industries, more and more networks and related applications with largescale have been constructed and implemented. Safe, efficient, and manageablenetworks and easy-to-maintain applications become more and more attractive byusers. Distributed Application System typically has tailored maintenance softwareto monitor and administrate the operating status at the initial construction period.However, those tailored maintenance software will be less effective once the entireSystem turns larger and larger, equipments becomes more and more, and function,amount and processing procedure needs more and more complex. The research ofthis project aims to improve the operational management on distributed applicationSystem with large scale.This article analyzes the distributed application System with large scale, whichhas a few characteristics including extensive users, multiple applications, multipleplatforms, flexibility and comprehensiveness etc. This research designed thecomprehensive solution for status monitoring System by analyzing its demands andcomparing the Strengths and weaknesses with other related products both fromoverseas and domestic. The System can monitor the status and collect the data ofoperating System, software and database, aslo it will alert when something goeswrong. At the same time the status monitoring System can be deployed by gradingand has the flexibility to be expanded.The flexible and dynamic collection and administration technology wasadopted by this research where users can collect customerlised monitoring items asper their requests. The alert correlation and origion tracing technology was alsoapplied to assist the system maintainer to catergarise and summarise the alertinformation then locate the fault point by virious menas including alert typecatergarilisation, causation leading rules, alert elimination rules etc. The interfaceprotocol was established by this research which enables the application programmeto imput the important information produced through internal operation to statusmonitoring system via interface, hence clearly and accurately acknowledge theoperation status of administrated application programme. Based on the collecteddata transmission of UDP protocol, the research realized the subcontractingtechnology of mass data transmission and end-to-end confirmation mechanism ofdata package, which satisfied the accuracy requirement of data transmission undercomplex network environment and unstable datas.The design and implementation of status monitoring System on distributedapplication in this research has reached related requirements through functional and performance tests. The satisfactory results have been achieved during theapplication process at National Computer Security Centre. With the improveness offunction postprocess, the System performance will be getting better and moredistuibuted application Systems will be utilized.