Author(s)

Fontaine Rafamantanantsoa^*, Maherindefo Laha

University of Fianarantsoa, Fianarantsoa, Madagascar.

The purpose of this study is to analyze and then model, using neural network models, the performance of the Web server in order to improve them. In our experiments, the parameters taken into account are the number of instances of clients simultaneously requesting the same Web page that contains the same SQL queries, the number of tables queried by the SQL, the number of records to be displayed on the requested Web pages, and the type of used database server. This work demonstrates the influences of these parameters on the results of Web server performance analyzes. For the MySQL database server, it has been observed that the mean response time of the Web server tends to become increasingly slow as the number of client connection occurrences as well as the number of records to display increases. For the PostgreSQL database server, the mean response time of the Web server does not change much, although there is an increase in the number of clients and/or size of information to be displayed on Web pages. Although it has been observed that the mean response time of the Web server is generally a little faster for the MySQL database server, it has been noted that this mean response time of the Web server is more stable for PostgreSQL database server.

Web Server, Performances, Neural Network, Database Server, MySQL, PostgreSQL, Apache, Mean Response Time, Apache Benchmark, SNNS

Rafamantanantsoa, F. and Laha, M. (2018) Analysis and Neural Networks Modeling of Web Server Performances Using MySQL and PostgreSQL. Communications and Network, 10, 142-151. doi: 10.4236/cn.2018.104012.