Calculating Software's Energy Use and Carbon Emissions: A Survey of the State of Art, Challenges, and the Way Ahead

2025-06-11

AI Summary

The escalating energy consumption and carbon emissions of software and AI systems necessitate rigorous measurement methodologies. Method: This paper systematically surveys and evaluates existing energy and carbon measurement approaches, proposing the first unified taxonomy classifying methods into monitoring-, estimation-, and black-box-based categories. It conducts a multidimensional assessment across hardware components (CPU, GPU, RAM) and dual dimensions—energy consumption and carbon emissions—grounded in bibliometric analysis and functional comparison of 87 tools. Contribution/Results: Key gaps are identified, including inadequate GPU dynamic power modeling and insufficient carbon intensity mapping for cloud environments. Three pervasive challenges are revealed: poor reproducibility, high hardware heterogeneity, and ambiguous system boundary definitions across the software lifecycle. The findings provide theoretical foundations and practical pathways for establishing standardized benchmarks and advancing green software engineering.

Technology Category

• Machine Learning: Efficient ML / Green AI
• Application Domains: Software Engineering
• Constraint Satisfaction and Optimization: Solvers and Tools

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Sustainability and carbon-aware systems for Web, mobile, and WoT
• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

The proliferation of software and AI comes with a hidden risk: its growing energy and carbon footprint. As concerns regarding environmental sustainability come to the forefront, understanding and optimizing how software impacts the environment becomes paramount. In this paper, we present a state-of-the-art review of methods and tools that enable the measurement of software and AI-related energy and/or carbon emissions. We introduce a taxonomy to categorize the existing work as Monitoring, Estimation, or Black-Box approaches. We delve deeper into the tools and compare them across different dimensions and granularity - for example, whether their measurement encompasses energy and carbon emissions and the components considered (like CPU, GPU, RAM, etc.). We present our observations on the practical use (component wise consolidation of approaches) as well as the challenges that we have identified across the current state-of-the-art. As we start an initiative to address these challenges, we emphasize active collaboration across the community in this important field.

Problem

• Measure software and AI energy and carbon emissions
• Compare tools for energy and carbon monitoring
• Address challenges in current emission measurement methods

Innovation

• Surveying software energy measurement methods
• Categorizing approaches as Monitoring, Estimation, Black-Box
• Comparing tools by components and granularity