Performance Testing And Optimization Of Performance Management System Based On The Integrated Telecommunication Network

In recent years, with the increase of traffic types and number of telecommunications network management, network is becoming more and more enormous and complex. Meanwhile, we pay more attention for the performance of network management. In this era. we save time as gold and focus on the user experience, the performance of system will directly influence the quality of service for enterprise promised to offer. Therefore, the performance testing and optimization of software gradually become very important. Looking for performance bottlenecks and optimizing performance are two significant goals for performance testing. How to analyze the system bottlenecks and how to optimize system performance are difficulties for performance testing, because the system performance bottlenecks may exist in large range. It is inevitable to test the performance of performance management system based on the massive integrated telecommunication network, in order to avoid serious consequences which caused by performance before used.This paper is based on a practical project of the performance management system of Alcatel lucent integrated telecommunication network. We focus on the system program structure, software testing theory and system performance bottlenecks. Firstly, we introduce the main structure and component modules of performance management system. Then, we summarize a testing and optimizing process which is suit for tested system. A series of performance testing and optimization are processed, according to these seven stages, which are:test scheme, test environment preparation, test scenario development, test execution, test results analysis and find bottlenecks, application program optimization, and regression test.According to the characteristics of the system, this paper designed and implemented a PM Data Simulation to produce performance Data file, and Java programs to produce corresponding data in database table. In addition, we improved the method of Rapid Bottleneck Identify, and adopted an approach which combines "experts consultation method" and "method for decomposition" to analyze performance bottlenecks in the system modules, such as Storage and Sch-Report modules. Then, we optimized the application based on the performance bottlenecks. In order to eliminate system bottlenecks and effectively improve the system performance, we retested the system after optimized. In addition, the JProfile performance testing tool is used to monitor the performance of Sch-Report module.