Model transformation chains in model-driven performance engineering: experiences and future research needs

Abstract

We gained experiences in implementing rule based model transformations within an industrial case study called Model-Driven Performance Engineering (MDPE). Similar to other MDE scenarios, these transformations have been implemented via multiple transformation steps interconnected in an automated model transformation chain. In this short paper, we use the MDPE case study to demonstrate reasons for decomposing model transformations and discuss disadvantages in terms of execution costs. Based on these experiences, we propose, as an input for future research, an architecture to optimize decomposed model transformation chains.