Embedding Energy Efficient Computing Practices

  • Al Daas, Hussam (STFC)
  • Huntley, Jessica (STFC)
  • Greenbank, Adam (STFC)

Please login to view abstract download link

ABSTRACT It has been estimated IT sector accounted for 2-6% of global carbon emissions in 2020 [1,2,3] and is predicted to rise to as much as 20% by 2030 [3]. Data centres alone are estimated to have emissions comparable to American commercial aviation [4]. The growing scale of computing systems have made energy-efficiency a central concern, creating an urgent need for new algorithms that are designed not only for performance and accuracy, but also for minimal energy consumption. Energy-efficient algorithms can significantly reduce power usage by optimising computation, memory access, communication, and data movement, which are often the dominant sources of energy cost. To guide and validate such designs, robust approaches and tools for measuring energy consumption and the associated carbon footprint are essential; these include hardware-level power sensors, software profilers, energy models, and life-cycle assessment methods that translate energy use into environmental impact. Benchmarking plays a critical role by providing standardized workloads, metrics, and experimental methodologies that enable fair comparison across algorithms, systems, and platforms, ensuring that energy efficiency claims are measurable and reproducible. These elements are deeply intertwined: accurate measurement tools inform meaningful benchmarks, benchmarks reveal algorithmic and system-level inefficiencies, and insights from both drive the design of new energy-aware algorithms. Together, they form a feedback loop that is fundamental to developing holistic approaches for energy-efficient computing, where algorithm design, measurement, carbon awareness, and benchmarking co-evolve to achieve sustainable computational performance. This talk will provide a summary of existing sustainable computing initiatives, application agnostic tools and resources, as well as approaches for embedding energy-efficient practices across research communities. REFERENCES [1] Malmodin, J., & Lundén, D. (2018). The Energy and Carbon Footprint of the Global ICT and E&M Sectors 2010–2015. Sustainability, 10(9), 3027. https://doi.org/10.3390/su10093027 [2] Belkhir, L., & Elmeligi, A. (2018). Assessing ICT global emissions footprint: Trends to 2040 & recommendations. Journal of Cleaner Production, 177, 448–463. https://doi.org/10.1016/j.jclepro.2017.12.239 [3] Andrae, A. S. G., & Edler, T. (2015). On Global Electricity Usage of Communication Technology: Trends to 2030. Challenges, 6(1), 117-157. https