📄 Spatio-Temporal Shifting to Reduce Carbon, Water, and Land-Use Footprints of Cloud Workloads

Abstract

In this journal article, we investigate the potential of spatial and temporal workload shifting to reduce the carbon, water, and land-use footprints of cloud computing workloads. Using real-world big data and FaaS workload traces, public data on AWS and Azure infrastructure, and available electricity grid data, we evaluate the impacts of spatio-temporal shifting decisions. Our results show that both temporal and spatial shifting can substantially reduce all three footprints, achieving reductions of up to 85% in operational carbon emissions, around 50% in water consumption, and up to 45% in land use, with spatial shifting providing the largest benefits.