CarbonSim: A Lifecycle-Aware Framework for Evaluating Carbon Tradeoffs in Hardware Upgrade Decisions
2026-06-04 • Distributed, Parallel, and Cluster Computing
Distributed, Parallel, and Cluster ComputingOperating Systems
AI summaryⓘ
The authors created CarbonSim, a tool that helps figure out the overall carbon emissions involved when deciding to upgrade computer hardware. They show that newer, more energy-efficient machines aren't always better for the environment if the old ones can still be used, especially when electricity comes from clean sources or computers aren't working hard. Their tool looks at how much energy the computers use, the carbon cost to make them, and how clean the electricity is over time. This means decisions about upgrading should think about how the computers are used, where they are, and their full environmental impact.
carbon emissionslifecycle analysishardware upgradeenergy efficiencyworkload utilizationgrid carbon intensityembodied carbonscheduling policiesICT environmental impact
Authors
Kartik Hans, Kaiwen Zhao, Stephen Lee
Abstract
As the demand for information and communication technologies (ICT) continues to rise, the environmental impact of computing systems is becoming an increasingly critical concern. Although newer hardware often improves performance and energy efficiency, these gains do not always offset the carbon cost of premature replacement, particularly under low-utilization workloads or low-carbon electricity grids. We present CarbonSim, a lifecycle-aware simulation framework for evaluating carbon tradeoffs in hardware upgrade decisions. CarbonSim combines workload execution profiles, machine-level power characteristics, embodied carbon inventories, scheduling policies, and time-varying grid carbon intensity to estimate total emissions under alternative deployment scenarios. The framework supports multiple embodied-carbon accounting strategies, including uniform amortization and front-loaded lifecycle attribution, enabling analysis under different hardware lifespan assumptions. Using heterogeneous CPU generations as calibration platforms, we demonstrate that newer machines do not always minimize total emissions: under lightly loaded workloads or cleaner electricity mixes, extending the useful life of existing hardware can reduce lifecycle carbon despite lower operational efficiency. These results highlight that hardware refresh decisions should be workload-aware, location-aware, and lifecycle-aware.