Not All Water Consumption Is Equal: A Water Stress Weighted Metric for Sustainable Computing
This work addresses water sustainability in computing for researchers and industry, offering a novel metric to guide decisions, though it is incremental in refining existing water impact assessments.
The authors tackled the problem of assessing water consumption in computing by accounting for spatial and temporal water stress variations, resulting in the SCARF framework that calculates an Adjusted Water Impact metric and demonstrates opportunities for reducing water impact through optimized location and time choices in case studies on LLM serving, datacenters, and semiconductor fabrication.
Water consumption is an increasingly critical dimension of computing sustainability, especially as AI workloads rapidly scale. However, current water impact assessment often overlooks where and when water stress is more severe. To fill in this gap, we present SCARF, the first general framework that evaluates water impact of computing by factoring in both spatial and temporal variations in water stress. SCARF calculates an Adjusted Water Impact (AWI) metric that considers both consumption volume and local water stress over time. Through three case studies on LLM serving, datacenters, and semiconductor fabrication plants, we show the hidden opportunities for reducing water impact by optimizing location and time choices, paving the way for water-sustainable computing. The code is available at https://github.com/jojacola/SCARF.