The Design And Implementation Of Blog Management Module In Web Server Performance Analysis Tool

With the development and wide application of information technology, especially the rapid spread of mobile technology and cloud computing, the requirement of web engineering development server performance has become more and more important. Although there are a lot of testing and performance analysis tools in the field of development, any of them doesn’t meet some of the requirements, such as analysis of the historical performance of each server data, server-side code for each module performance data, as well as project management combined by the real-time performance analysis to the effective management of the development process.During the author’s internship, the project team’s jobs are the Mobile project development and Web development. These projects are related to the observation and analysis of the problem of server performance, that is how to get the server performance data after the developers code has been automatically submitted for this version, and inform developers. Development team requires a scalable, highly portable, lightweight tool to accomplish this quest..The thesis has designed a log-based analysis of Web Server performance observation and statistical analysis tools, combined with Jenkins code version management and continuous integration, we use Hadoop theory and technology to support the log file storage, handling and statistics. In addition, the tool has considered scalability and portability logically, it has an SDK to include the core module package which facilitate the need for performance analysis code module to invoke, and use Hadoop cluster scalability to facilitate the test server cluster expansion.The scheme we propose is suitable for the observation and analysis of multi-server cluster on server performance, and the development using continuous integration tools for management. With the high ability of cluster storage and computing, the proposed scheme improves the ability of log files’ storage and statistical computing. It is suitable to the analysis of the code performance in the development of the project with different versions.