Abstract
This paper proposes that the same application running in a Platform as a Service
environment, which provides the same performance, will behave differently when using
different code bases. The same website has been built on top of four different coding
languages, Python, PHP, DotNet, and NodeJS, and hosted using the same environment
meaning there is no difference in computing power, location, network, or any other physical
difference one might experience within multiple on-premise datacentres. To measure the
performance, a benchmarking tool named JMeter has been used to run load tests over a
period of 30 days. The results have shown that PHP had the best score of 3.58 on a 5 Point
Likert Scale used in combination with a Weighted Matrix. Future studies should focus on
more complex applications and use other benchmarking tools to further evaluate the
performance within PaaS.
