Versatility or validity: A comprehensive review on simulation of Datacenters powered by Renewable Energy mix

Nowadays, DataCenters powered by Renewable Energy mix (DCRE) appear worldwide. It brings a hot topic on studying how to fully utilize renewable energy and save brown energy. However, for the vast and complex infrastructure, researchers would perform their studies on a DCRE simulator rather than a real-world DCRE. It brings a question: do these simulators have the same functions and result for the same optimization approach? This paper focuses on the question and proposes the versatility and validity models and relationships for these simulators. We use the term ”simulators” broadly, encompassing any general-purpose simulation tools, experiment-purpose simulation approaches, and other simulation works on DCREs. We have reviewed the simulation works for DCRE in the last ten years. After that, we rebut the widespread practice that a simulator has functions for the particular study with an assumed validity. Furthermore, we claim that uniform metrics for both versatility and validity lead to a better simulator. Finally, we back up this claim with comprehensive benchmarking. The experimental results show that the more versatile a simulator is, the better its validity.