Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Carbon Footprint Evaluation of Code Generation through LLM as a Service

Tina Vartziotis, Maximilian Schmidt, George Dasoulas, Ippolyti Dellatolas, Stefano Attademo, Viet Dung Le, Anke Wiechmann, Tim Hoffmann, Michael Keckeisen, Sotirios Kotsopoulos

TL;DR

The paper investigates the environmental footprint of AI-driven code generation and proposes a lifecycle-based framework to quantify embodied and operational carbon in LLM-based software workflows. It distinguishes two main lifecycles—the Model Development Cycle and the LLM as a Service (LLMaaS) Cycle—and introduces a Green Capacity metric and a Carbon Footprint taxonomy. Through a case study using GitHub Copilot in a software testing scenario, it presents embodied and operational emissions and demonstrates how green coding practices affect the overall footprint. The authors call for robust sustainability metrics and industry-specific evaluations (notably automotive) to guide greener AI-assisted development.

Abstract

Due to increased computing use, data centers consume and emit a lot of energy and carbon. These contributions are expected to rise as big data analytics, digitization, and large AI models grow and become major components of daily working routines. To reduce the environmental impact of software development, green (sustainable) coding and claims that AI models can improve energy efficiency have grown in popularity. Furthermore, in the automotive industry, where software increasingly governs vehicle performance, safety, and user experience, the principles of green coding and AI-driven efficiency could significantly contribute to reducing the sector's environmental footprint. We present an overview of green coding and metrics to measure AI model sustainability awareness. This study introduces LLM as a service and uses a generative commercial AI language model, GitHub Copilot, to auto-generate code. Using sustainability metrics to quantify these AI models' sustainability awareness, we define the code's embodied and operational carbon.

Carbon Footprint Evaluation of Code Generation through LLM as a Service

TL;DR

The paper investigates the environmental footprint of AI-driven code generation and proposes a lifecycle-based framework to quantify embodied and operational carbon in LLM-based software workflows. It distinguishes two main lifecycles—the Model Development Cycle and the LLM as a Service (LLMaaS) Cycle—and introduces a Green Capacity metric and a Carbon Footprint taxonomy. Through a case study using GitHub Copilot in a software testing scenario, it presents embodied and operational emissions and demonstrates how green coding practices affect the overall footprint. The authors call for robust sustainability metrics and industry-specific evaluations (notably automotive) to guide greener AI-assisted development.

Abstract

Due to increased computing use, data centers consume and emit a lot of energy and carbon. These contributions are expected to rise as big data analytics, digitization, and large AI models grow and become major components of daily working routines. To reduce the environmental impact of software development, green (sustainable) coding and claims that AI models can improve energy efficiency have grown in popularity. Furthermore, in the automotive industry, where software increasingly governs vehicle performance, safety, and user experience, the principles of green coding and AI-driven efficiency could significantly contribute to reducing the sector's environmental footprint. We present an overview of green coding and metrics to measure AI model sustainability awareness. This study introduces LLM as a service and uses a generative commercial AI language model, GitHub Copilot, to auto-generate code. Using sustainability metrics to quantify these AI models' sustainability awareness, we define the code's embodied and operational carbon.

Paper Structure

This paper contains 14 sections, 4 equations, 1 figure, 4 tables.

Table of Contents

  1. Introduction
  2. Impact of AI on the Environment
  3. LLMs for sustainable code generation
  4. Green Coding Practices
  5. Carbon Footprint Taxonomy
  6. Model Development Cycle.
  7. LLM as a Service Cycle.
  8. Embodied Costs
  9. Case Study in Software Testing
  10. Operational costs
  11. Sustainability Metrics for Software Operation
  12. Operational Costs of Case Study in Software Testing
  13. Conclusion
  14. Acknowledgments

Figures (1)

  • Figure 1: The cycle phases where operational and embodied carbon footprints occur. We study two production cycles: i) the Model Development Cycle (left), and ii) the LLMaaS Cycle (right). The embodied carbon footprint of the LLMaaS Cycle is overlapping in the LLM inference phase with the operational one by the Model Development Cycle.