Towards a Domain-Specific Language for Provisioning Multiple Cloud Testing Environments for Mobile Applications

Abstract

Towards a Domain-Specific LanguagTowards a Domain-Specific Languag Provisioning testing environments for mobile applications is one of the most significant challenges within the software industry. Due to this high complexity that exists when provisioning test environments within the multiple available cloud platforms, it is necessary to make a significant investment in human resources, like time and effort for the implementation and execution of testing. There is additional complexity: testing software in a single environment is no longer sufficient. Today's mobile industry is constantly growing, and execution environments tend to be always different; the hardware configuration is usually different and sometimes exceeds the software barrier. It is challenging to execute testing on each of the existing devices, as this requires a long task of human intervention. Today some platforms provide testing services in different environments. However, not all providers have the complete set of environments that one would like to have, and specific knowledge is mandatory for using each available tool. It is a task that requires expertise and time. This work seeks to mitigate the impact on time and the learning curve through a high-level tool developed using a model-oriented approach, thus reducing the time needed for setting up each required platform for organizations. As a solution, we propose a Domain-Specific Language for provisioning multiple cloud testing environments for mobile applications. The configuration of the environment is done with the Domain Specific Language to make the usage easier by the final user. The necessary code is generated through transformations to set up an environment on cloud platforms such as Amazon Web Services (AWS) and Google Cloud Platform (GCP). This usage of this platform results in fewer code lines written and less time learning about the specific knowledge for each platform.