Design And Implementation Of Operation And Maintenance Management Platform For Teaching Live Broadcast System

Aiming at the requirements of high timeliness in operation and maintenance management,in order to improve the efficiency of operation and maintenance management and the utilization of resources such as servers,a set of operation and maintenance management platform for teaching live broadcast system is designed and implemented in this thesis.The flowing researches have been executed for the requirements of this platform in high compression capacity,large data storage and improving resource utilization.1)The design of operation and maintenance data collect SDK and report protocol.A set of operation and maintenance data collection SDK is designed in this thesis.The operation and maintenance data such as CPU and memory utilization rate,network sending and receiving traffic,online status,and service status for business system were collected by this SDK.An unified data report protocol is further designed with the principle of saving data report time and improving efficiency in order to complete the data report collected by above SDK.2)The design of platform architecture and data storage middleware.Firstly,the platform architecture is designed by SOA according to some requirements such as high compression capacity,large data storage and high timeliness.Then a data storage middleware is designed according to data storage requirements.The MySQL database management system is used to carry out the storage and management of basic data in this platform.MongoDB is used to carry out the storage and management of historical data in this platform.Redis is used to complete the real-time data cache management to improve the real-time data access efficiency.Finally,we designed a set of interfaces for this platform according to the demand analysis.This interface is realized by SSM framework.3)The design of resource scheduling and scheme evaluating model.A set of resource schedule and scheme evaluating model is designed according to the historical operation and maintenance data in this thesis for the purpose of improving server resource utilization,and the scheduling of applications could be achieved by this model.Experiments are conducted for the effectiveness of operation and maintenance data collect and report,the performance of platform interface and data storage,and the feasibility of resource scheduling model in this thesis.The experimental results have shown that the operation and maintenance data collection is accurate and the data report has an advantage of low latency.And we can draw some conclusions that our platform runs smoothly,the interface works stably,and the data storage middleware could better meet the requirements of effectiveness in the operation and maintenance of our business system through the experiments of platform performance such as compression and high concurrency.The test of our resource scheduling model has shown that the model has certain rationality and feasibility to a certain extent.