The Jevons Paradox In Cloud Computing: A Thermodynamics Perspective

TL;DR

This thermodynamic model provides energy consumption insights into modern hyperscale clouds, and it is found that system growth, due to the revenue generation of cloud platforms, is a key driver behind energy consumption.

Abstract

How do we explain the simultaneous growth in energy efficiency of cloud computing and its energy consumption? The Jevons paradox provides one perspective of this phenomenon. However, it is not clear or obvious \emph{why} the Jevons paradox exists, and \emph{when} is it applicable. To answer these questions, we seek inspiration from thermodynamics, and model the cloud as a thermodynamic system. We find that system growth, due to the revenue generation of cloud platforms, is a key driver behind energy consumption. This thermodynamic model provides energy consumption insights into modern hyperscale clouds, and we validate it using data from Meta and Google. Our investigation points to the necessity of future work in new and meaningful efficiency metrics, implications for future applications and edge clouds, and the need for studying system-wide energy and sustainability.

Figures (6)

• Figure 1: Energy consumption of data centers in the US and the world is increasing, inspite of all efficiency advances. Data from mckinsey_dciea24shehabi2016united.
• Figure 2: Thermodynamic model of cloud computing. Energy and matter flow from the higher-potential reservoir. Work (revenue) expands the system's interface and increases the energy consumption further.
• Figure 3: Data from Meta and Google shows that energy use is proportional to the cumulative revenue, which is the central prediction of the thermodynamics model (dashed lines).
• Figure 4: Embodied emissions can be a good proxy for system size, and have a constant relation with the cumulative energy use, as predicted by the model.
• Figure 5: The Capital Expenditure (CapEx) used for building data centers can also be used as a proxy for system size.