publications

2025

1. Metrics for Data Center Embodied Carbon

   Leo Han, Yueying Lisa Li, and Udit Gupta

   SIGMETRICS Perform. Eval. Rev., Aug 2025

   Abs DOI

   The information and computing technology (ICT) industry accounts for a significant and growing portion of total global carbon emissions: composing 2.1% to 3.9% of global emissions in 2021 with emissions growing annually [7]. To address these emissions, Microsoft [16] and Google [14] have each pledged to reach carbon neutrality or net-zero carbon by 2030, and Amazon [17] has pledged to achieve the same by 2040. The push towards sustainable data centers has led to a growing emphasis on incorporating carbon as part of optimization metrics across the computing stack including systems [11], architecture [8, 4, 5, 6], and software [18, 10].

2. Fair, Practical, and Efficient Carbon Accounting for LLM Serving

   Yueying Lisa Li, Leo Han, G Edward Suh, Christina Delimitrou, Fiodar Kazhamiaka, Esha Choukse, Rodrigo Fonseca, Liangcheng Yu, Jonathan Mace, and Udit Gupta

   SIGMETRICS Perform. Eval. Rev., Aug 2025

   Abs DOI

   We propose a framework for evaluating carbon attribution methods for multi-tenant LLM serving. The framework formalizes the problem using three key components: (1) a set of requests with varying prompt and decode lengths, (2) the LLM inference runtime including batching algorithms, and (3) a carbon emission model accounting for both operational carbon (proportional to power consumption and carbon intensity) and embodied carbon from hardware manufacturing. Using the Shapley value as ground truth for fair attribution, we demonstrate why simple ’leave-one-out’ attribution methods fail to satisfy efficiency properties. The framework evaluates attribution methods against four criteria: scalability (computational complexity), fairness (minimizing deviation from Shapley values), sample efficiency (algorithmic approximations for complex cases), and incentivization (encouraging users to optimize their usage patterns).

3. Fair-CO2: Fair Attribution for Cloud Carbon Emissions

   Leo Han, Jash Kakadia, Benjamin C. Lee, and Udit Gupta

   In Proceedings of the 52nd Annual International Symposium on Computer Architecture, Aug 2025

   Abs DOI

   Fair-CO2 is a system for fairly attributing operational and embodied carbon in cloud data centers to user workloads. It leverages the Shapley value, a game theory solution for fair shared cost attribution with theoretical fairness guarantees. We propose the standard Shapley value solution as a ground truth for attribution in cloud data centers, addressing two key gaps in existing carbon attribution methods that lead to unfair attributions: the effect of dynamic demand on embodied carbon, and interference effects in colocated scenarios. However, the computational cost of the Shapley value solution scales exponentially with the number of workloads and becomes intractable for large systems. Fair-CO2 addresses the scalability challenges of the Shapley value while preserving its fairness benefits via two core components: demand-aware embodied carbon attribution and interference-aware resource cost attribution. Using Monte Carlo simulations of workload schedules and colocation scenarios, we show that Fair-CO2 can approximate the ground truth Shapley attribution solution at scale. We also show how users, once provided a fair way of estimating their workload carbon footprint, can dynamically optimize workload deployment for carbon savings.

4. CarbonClarity: Understanding and Addressing Uncertainty in Embodied Carbon for Sustainable Computing

   Xuesi Chen, Leo Han, Anvita Bhagavathula, and Udit Gupta

   Aug 2025