A study on the improvement of supercomputer energy efficiency based on green500 benchmarking data

Abstract

With the rapid growth of AI-related industries, the need for reducing and optimizing energy consumption in large-scale computational resources, such as supercomputers, has become increasingly important. This study focuses on supercomputers listed in the Green 500, categorizing existing benchmarking evaluation variables into input and output factors. An energy efficiency objective function was introduced, and DEA was conducted using BCC and SEM models. The study analyzed the relative efficiency levels among supercomputers and identified factors and levels of potential efficiency improvements. The results provide insights into the performance factors of individual supercomputers and their potential for improvement. Furthermore, by comparing the energy efficiency evaluated by the Green 500 with the results of DEA, it demonstrated the potential for utilizing DEA as a new means for efficiency improvement. It also highlighted the necessity of a comprehensive evaluation that incorporates various performance factors, rather than a simple efficiency assessment based solely on energy consumption.