|
|
2014 » Papers » Volume 1 » Performance time for e-learning applications with multiple databases 1. PERFORMANCE TIME FOR E-LEARNING APPLICATIONS WITH MULTIPLE DATABASES Authors: Truica Ciprian-Octavian, Boicea Alexandru, Radulescu Florin Volume 1 | DOI: 10.12753/2066-026X-14-017 | Pages: 121-128 | Download PDF | Abstract
In this paper we describe the use of multiple types of database management systems in the same e-learning application. We will present three web applications one that uses a PostgreSQL database management system, one that uses a MongoDB NoSQL management system and one that uses both. These applications have been developed using the MVC design pattern and will be used to store scientific articles. For the web application that uses the multiple database solution we will link the two databases in the controller module by using the primary keys in the PostgreSQL tables to correspond to primary keys in the MongoDB collections. We will also do stress test to determine the performance time for HTTP request so that we can conclude if it's better to use a multiple database approach over a RDBMS (Relational Database Management System) or NoSQL one. The web applications will be developed using the high-level Python Web framework Django. For a web server we have chosen Gunicorn (Green Unicorn), a pre-fork worker model Python WSGI (Web Server Gateway Interface) HTTP Server for UNIX. The Gunicorn web server will be configured to start with 20 workers so that it will accept 20 concurrent connections. The development, databases, web servers and tests have been done on an Ubuntu 12.04LST x64 operating system. The stress tests will use the httplib python module that implement the client side of the HTTP and HTTPS protocols and we will focus our attention on the read operation by testing for each application its HTTP response time. We have used thread to simulate concurrent access to the web server, each thread will use the GET method to request a page, wait for the page to load and then close the connection. | Keywords
NoSQL; RDBMS; web applications; web server; ACID; BASE |
|
|
|