A Framework for Evaluating Maintainability and Performance of Object-Relational-Mapping Tools in Web Application Frameworks

Abstract

Web Application Frameworks (WAF) and Object-Relational Mapping (ORM) tools reduce the time and effort needed to develop high quality applications. However, there are many of them and choosing the best suited for a particular project is not easy. There are comparative studies about these tools, inconveniently, the existing studies do not considerate a domain complex enough that allow us to measure in a more precise way their advantages and disadvantages, also they do not analyze the maintainability related to ORM usage. To contribute in understanding these tools, we propose a WAF-ORM evaluation framework that measures performance and maintainability which we used to compare a group of WAF-ORM technologies. Our study includes a case study that separates queries in three different complexity levels that use both pure SQL and ORM’s API. We addressed performance analysis by studying the impact of queries’ complexity and database size, leaving all other variables affecting performance constant. Also, as maintainability is harder to measure, we propose a qualitative way to classify WAF-ORM technologies. A comparison based on these aspects is important because understanding performance and architectural benefits of ORMs technologies is helpful for both users and developers of WAF-ORM technologies.