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

TL;DR

This study explores how spatial and temporal workload shifting in cloud computing can significantly reduce environmental footprints, demonstrating substantial savings in carbon, water, and land use through simulation of real-world data.

Contribution

It introduces a simulation-based approach to quantify environmental benefits of workload shifting strategies in cloud infrastructure, highlighting their effectiveness and robustness.

Findings

Spatial shifting can reduce carbon, water, and land footprints by up to 85%, 50%, and 45%.

Combined spatiotemporal shifting yields the greatest overall environmental reduction.

Shifting strategies are robust to prediction errors and seasonal variations.

Abstract

In this paper, we investigate the potential of spatial and temporal cloud workload shifting to reduce carbon, water, and land use footprints. Specifically, we perform a simulation study leveraging publicly available data on the cloud infrastructure of major providers (AWS and Azure) as well as real-world workload traces (big data analytics and FaaS) and grid mix data to consider two different scenarios. Our simulation results indicate that spatial shifting can substantially lower carbon, water, and land use footprints. In the FaaS applications, shifting the spatiotemporal workload achieves carbon savings of up to 85%, water savings of around 50%, and reductions in land use of up to 45%, all while optimizing for the respective factors. Mixed optimization yields results comparable to those of land use alone. For big data workloads, spatiotemporal shifting delivers reductions of up to 45% in carbon emissions, 40% in water consumption, and nearly 40% in land use when optimized for the respective factors. Temporal shifting also decreases the footprint, though to a lesser extent. When applied together, the two strategies yield the greatest overall reduction, driven mainly by spatial shifting with temporal adjustments providing an additional, incremental benefit. Sensitivity analysis demonstrates that such shifting is robust to prediction errors in grid mix data and to variations across different seasons.