The Design And Implementation Of Database For Network Monitoring System

In the information age, the development of the Internet needs more support from the database technology. Under the condition of the vast amounts of data information, users focus more attention on the stability and the efficiency of the solution of the database.In order to find out the stable and efficient database of monitoring the enterprise employees network behavior, this thesis is based on monitoring the network behavior of the enterprise employees and introducing the efficiency of the MySQL database. The MySQL database use the LAMP architecture to develop and run a Linux system to save the web log information. The design and the performance optimization of the MySQL database can process the read and write frequent large-scale logging data. The MySQL database can ensure the stable and efficient database of monitoring the enterprise employees network behaviorTo improve the efficiency of querying huge amounts of data, this thesis designs the data fractionation of the static data and dynamic data.That is, historical and real-time data.First,turn the dynamic data into the static data by making the task at regular time and then use a table partition technology to partition the data,and then use the memory table and disk tables to save the basic data which is frequently accessd but only small numbers. These steps are to improve the efficiency of the table connection. In monitoring strategy module, adopting the flexible control strategy table design,using the template form application strategy to the controlled object, and the application of set priority monitoring strategy are easy for administrators to operate the data. Design a set of efficient and reasonable database Schema.For optimizing the Query, this thesis analyzes the explain execution plan which occurs frequently in the Query statement and designs the appropriate indexes. The thesis also designs the classification data model instead of the less suitable for indexing columns of data. That is, design according to the data table column attributes. Classify the data based on one-dimensional or two-dimensional relationship and save for the view mode. Through the match of the corresponding subset of the data, it can implement the Query statement reorganization and replace the original Query statement.Test results show that the optimized Query, due to avoiding the delay of the full table scan response time, the efficiency of querying has been greatly improved. It proves the validity of the optimization scheme.