Search papers, labs, and topics across Lattice.
CarbonSim is a lifecycle-aware simulation framework designed to evaluate the carbon tradeoffs associated with hardware upgrade decisions in the context of ICT. By integrating workload execution profiles, machine-level power characteristics, and carbon intensity data, the framework reveals that newer hardware does not always lead to lower total emissions, particularly in scenarios with low-utilization workloads or cleaner electricity sources. The findings advocate for a more nuanced approach to hardware refresh decisions, emphasizing the importance of considering workload, location, and lifecycle factors to minimize carbon footprints.
Newer hardware can actually increase total emissions under certain conditions, challenging the conventional wisdom of frequent upgrades.
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.
