Modeling the Carbon Footprint of HPC: The Top 500 and EasyC

2025-09-16

AI Summary

Existing carbon footprint assessments of Top 500 supercomputing systems suffer from pervasive data scarcity and insufficient modeling granularity. Method: We propose the first scalable, system-wide carbon quantification framework, introducing EasyC—a lightweight tool that leverages minimal publicly available metrics (e.g., runtime power draw, hardware configuration) and integrates data interpolation with provenance-aware enhancement to achieve 98% system coverage. Contribution/Results: This work delivers the first comprehensive operational carbon (391 systems) and embodied carbon (283 systems) modeling for the entire Top 500 list, estimating annual operational emissions at 1.394 Gt CO₂e and embodied carbon at 1.882 Gt CO₂e. It further projects emission trends through 2030. The framework establishes a reproducible, extensible benchmark methodology and empirically grounded evidence base for green governance of high-performance computing infrastructures.

Technology Category

• Data Mining & Knowledge Management: Scalability, Parallel & Distributed Systems
• Machine Learning: Efficient ML / Green AI
• Humans and AI: Other Foundations of Human Computation & AI

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Sustainability and carbon-aware systems for Web, mobile, and WoT
• Responsible Web: Sustainability and climate impact of web technologies
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Climate change is a critical concern for HPC systems, but GHG protocol carbon-emission accounting methodologies are difficult for a single system, and effectively infeasible for a collection of systems. As a result, there is no HPC-wide carbon reporting, and even the largest HPC sites do not do GHG protocol reporting. We assess the carbon footprint of HPC, focusing on the Top 500 systems. The key challenge lies in modeling the carbon footprint with limited data availability. With the disclosed Top500.org data, and using a new tool, EasyC, we were able to model the operational carbon of 391 HPC systems and the embodied carbon of 283 HPC systems. We further show how this coverage can be enhanced by exploiting additional public information. With improved coverage, then interpolation is used to produce the first carbon footprint estimates of the Top 500 HPC systems. They are 1,393.7 million MT CO2e operational carbon (1 Year) and 1,881.8 million MT CO2e embodied carbon. We also project how the Top 500's carbon footprint will increase through 2030. A key enabler is the EasyC tool which models carbon footprint with only a few data metrics. We explore availability of data and enhancement, showing that coverage can be increased to 98% of Top 500 systems for operational and 80.8% of the systems for embodied emissions.

Problem

• Modeling carbon footprint of HPC systems with limited data
• Estimating operational and embodied emissions for Top500 supercomputers
• Projecting future carbon footprint growth through 2030

Innovation

• EasyC tool models carbon with few metrics
• Uses Top500 data and public information enhancement
• Interpolation estimates carbon footprint for HPC systems