Examining the Environmental Impact of High-Powered Computing
In parallel with the notable contributions of high-performance computing (HPC) to advancements in research, industry, and digital infrastructure, its surging energy demands and associated carbon footprint have emerged as pressing environmental concerns. Recent projections suggest that, without appropriate intervention, global electricity usage by data centers could surge by up to 160% by 2030, posing a major obstacle to achieving carbon neutrality. To confront this issue, this paper introduces a quantitative and layered assessment framework that evaluates the carbon emissions of HPC systems, factoring in operational intensity, regional deployment patterns, and power grid composition. Furthermore, it integrates water consumption analysis arising from cooling mechanisms to establish a dual-focus environmental indicator system that highlights both carbon and water-related impacts. Simulation results suggest that even under moderate usage rates (e.g., 56.25%), annual carbon emissions could exceed 290 million metric tons, and potentially reach over 500 million tons if energy structures worsen. Hence, accelerating the shift towards renewable and clean energy emerges as a vital trajectory for sustainable HPC development
References
- Shu, P., Chen, J., Liu, Z., Zhao, H., Li, X., & Liu, T. (2025). Survey of HPC in US Research Institutions. arXiv preprint arXiv:2506.19019.
- Adedoyin, F. F., Nwulu, N., & Bekun, F. V. (2021). Environmental degradation, energy consumption and sustainable development: accounting for the role of economic complexities with evidence from World Bank income clusters. Business Strategy and the Environment, 30(5), 2727-2740.
- Matthews, B., & Dall’Orsoletta, F. (2020). Extraction rates and the environmental impacts of economic growth in the twenty-first century. In Decent Work and Economic Growth (pp. 414-426). Cham: Springer International Publishing.
- Gołaś, Z. (2023). Decoupling analysis of energy-related carbon dioxide emissions from economic growth in Poland. Energies, 16(9), 3784.
- Ahmed, K., Liu, J., & Wu, X. (2017, September). An energy efficient demand-response model for high performance computing systems. In 2017 IEEE 25th international symposium on modeling, analysis, and simulation of computer and telecommunication systems (MASCOTS) (pp. 175-186). IEEE.
- Lannelongue, L., Grealey, J., & Inouye, M. (2021). Green algorithms: quantifying the carbon footprint of computation. Advanced science, 8(12), 2100707.
- Van Schie, F. T., Zavřel, V., Torrens, J. I., Hundertmark, T. F. W., & Hensen, J. L. (2015, December). Optimizing the total energy consumption and CO2 emissions by distributing computational workload among worldwide dispersed data centers. In 14th Conference of International Building Performance Simulation Association, BS 2015 (pp. 1118-1125). International Building Performance Simulation Association (IBPSA).
- Lannelongue, L., Grealey, J., & Inouye, M. (2020). Green Algorithms: Quantifying the carbon emissions of computation, arXiv. arXiv preprint arXiv:2007.07610.