Research On Optimization Method Of Fuel Information Database Performance

Database technology is the core of computer data processing and information management system, with the continuous expansion of the field of database usage and scale, the performance of database system has become more and more serious, it is a great challenge to ensure the database application system work normally and efficiently. Therefore, the research of database performance optimization has very important significance for improving the efficiency of the application system.After analyzing the research status at home and abroad of database performance optimization, in view of the performance problems that may arise in the long-term operation of the fuel information system, two types of database performance tuning and optimization schemes are proposed. The first scheme is to adjust the parameters of the database memory, the memory parameters which need to adjust is determined by analyzing the database access methods and the impact of caching on the execution of SQL statements, the mathematical model is established on the memory parameters and database performance, conic fitting algorithm and the greedy algorithm are adopted to get the optimal value of each parameter in turn, and through the local optimum to get the global optimal is used finally. The second scheme is to partition table which may need to be partitioned. Based on the storage characteristics of the data in the table, partition methods is selected, according to the structure of the table and the characteristics of the query on the table, table partition strategy is developed which can improve database performance, and then the table partitioning is implemented by the program.In the end, a database performance monitoring system has been designed and implemented, four kinds of performance information of the database are displayed in the system with the form of line chart. The adjustment of the parameters of the database memory has been supported through the graphical interface, and the computer program is used to build mathematical model and solve the local optimum of the memory parameter values. Meanwhile, the system has implemented the partition of the table and supported batch table partition operation and recommended to the user for all suitable partition tables in accordance with the partition strategy.