The Performance Analysis And Optimization Of The Police Performance Appraisal System

As the Web technology continues to mature and development, Web systems havebeen widely used in the field of government management and office, and that ise-government. It has become essential tools of government efficient office. To theWeb e-government system which has a massive computing need, there is an urgentneed to find a suitable solution to improve the system computing performance,shorten the response time and improve the efficiency of government work.The number of users and business data of the Police performance appraisalsystem is very large, which covers Tianjin Urban, subordinate branch, police station,as well as all levels of offices. This paper analyzes the bottleneck of the system in thecalculation speed and data access through testing the key performance indicator suchas response time by using Web system evaluation techniques and improves thecomputing power of the Police performance appraisal system from four aspects. First,set recalculation flag in the data table. By using this method, only data which aremodified will be recalculated after modifying data. Because few data are modified, sothe calculation will be greatly reduced in the process of recalculation. Second, take thehash table cache mechanism. Put the large amount of data in the database into hashtable before calculating, thus long waiting time produced by frequent access to thedatabase will be avoided. Third, use multi-threaded parallel computing. The Webserver has a multi-core CPU and the data need to calculate have a high degree ofparallelism so taking full advantage of the computing resources can greatly improvethe computing speed. Fourth, take a scores caching mechanism. Give the calculationwork to the administrator. When users click on the scores view page, read directly thescores which has already been calculated and stored in the database by theadministrator. Thus, the response time will be shortened.These four improving system performance method will be discussed in detail anddemonstrated by the experimental data in this paper and ultimately realized throughactual coding. Finally the computing power of the Police performance appraisalsystem has been greatly improved, user response time has been greatly reduced,thereby the efficiency of the Public security bureau officers has been increased, the operating costs have been reduced, and the effectiveness of the police managementmechanism has got a qualitative leap.