智算中心论文专站

LLM-Powered Predictive Decision-Making for Sustainable Data Center Operations

Hanzhao Wang, Jingxuan Wu, Yumeng Li, Yu Pan, Guanting Chen

Published 2026-08-19 · arXiv · Credibility S

The growing demand for AI-driven workloads, particularly from Large Language Models (LLMs), has raised concerns about the significant energy and resource consumption in data centers. This work introduces a novel LLM-based predictive scheduling system designed to enhance operational efficiency while reducing the environmental impact of data centers. Our system utilizes an LLM to predict key metrics such as execution …

Abstract, interpretation and reference

Abstract

The growing demand for AI-driven workloads, particularly from Large Language Models (LLMs), has raised concerns about the significant energy and resource consumption in data centers. This work introduces a novel LLM-based predictive scheduling system designed to enhance operational efficiency while reducing the environmental impact of data centers. Our system utilizes an LLM to predict key metrics such as execution time and energy consumption from source code, and it has the potential to extend to other sustainability-focused metrics like water usage for cooling and carbon emissions, provided the data center can track such data. The predictive model is followed by a real-time scheduling algorithm that allocates GPU resources, aiming to improve sustainability by optimizing both energy consumption and queuing delays. With fast inference times, the ability to generalize across diverse task types, and minimal data requirements for training, our approach offers a practical solution for data center scheduling. This framework demonstrates strong potential for advancing sustainability objectives in AI-driven infrastructure. Through our collaboration with a data center, we achieved a 32% reduction in energy consumption and a 30% decrease in waiting time.

Steady-State Equivalent Circuit Model for Data Center Loads

Muhammad Hamza Ali, Peng Sang, Hyeon Woo, Hyein Kang, Sungyun Choi, Amritanshu Pandey

Published 2026-08-18 · arXiv · Credibility S

Planners currently represent data centers as aggregate constant-PQ or ZIP loads in steady-state interconnection and contingency studies. These aggregate models are computationally convenient. However, they obscure the electrical relationship between computational workloads, server utilization, and grid-side demand. They ignore the internal power-electronic conversion stages of IT loads and assume homogeneous workloa…

Abstract, interpretation and reference

Abstract

Planners currently represent data centers as aggregate constant-PQ or ZIP loads in steady-state interconnection and contingency studies. These aggregate models are computationally convenient. However, they obscure the electrical relationship between computational workloads, server utilization, and grid-side demand. They ignore the internal power-electronic conversion stages of IT loads and assume homogeneous workload distributions across the compute clusters. This hides operating-point-dependent converter losses and efficiency variations. We propose a steady-state equivalent-circuit model (ECM) for data centers, which explicitly builds circuit models for IT loads, power supply units, cooling, and auxiliary systems. For power supply units, the equivalent circuit model explicitly represents internal power-electronic conversion stages. For IT loads, we develop a utilization-dependent server power model, and we combine it with loss-aware ECMs of power supply units. This approach captures the grid-side impact of heterogeneous workload distributions while preserving compatibility with conventional power-flow analysis. We evaluate this data center ECM in large-scale transmission power flows, using Monte Carlo simulations under heterogeneous and homogeneous cluster utilization. In comparison with the fixed-efficiency constant-PQ model, the ECM predicts that the most stressed line exceeds its thermal limit in about 30% of Monte Carlo samples. The results further show that homogeneous server utilization overstates line-loading variability by 17%-46% relative to heterogeneous server utilization, depending on the intra-cluster workload correlation.

Beyond the Grid: Cost, Carbon, and Capital Requirements of On-Site Power Technologies for AI Data Centers

Eliseo Curcio

Published 2026-08-08 · arXiv · Credibility S

Interconnection queues, not electricity prices, now govern where data centers can be built, and the standard levelized-cost comparison answers a question no developer faces: it assumes a load profile, freezes the grid price while modeling the demand that moves it, and quotes busbar costs a facility cannot buy. This paper evaluates nine on-site supply technologies against a delivered grid whose price is endogenous to…

Abstract, interpretation and reference

Abstract

Interconnection queues, not electricity prices, now govern where data centers can be built, and the standard levelized-cost comparison answers a question no developer faces: it assumes a load profile, freezes the grid price while modeling the demand that moves it, and quotes busbar costs a facility cannot buy. This paper evaluates nine on-site supply technologies against a delivered grid whose price is endogenous to projected data-center demand, on a complete-site basis that retains standby charges, with measured GPU training load, delivered fuel prices, production-pathway carbon, and statutory 45V and 48E incentive mechanics. Nothing beats the wire: gas combined cycle produces at 47 USD/MWh but costs about 114 USD per megawatt-hour of complete site energy against a 92 USD grid; four-hour storage is physically capped near 18 percent of annual energy and, charged at the margin, dirtier than the grid; hydrogen from grid-priced power fails on cost and carbon together. An investment inversion converts these findings into capital terms: conversion-hardware learning buys nothing, because free hardware still exceeds the grid for every low-carbon arm, while global electrolyser deployment on sited sub-20 USD/MWh power brings PEM hydrogen power to about 2.2 times the grid at 300 billion USD and 1.9 times at 1 trillion USD (2.7 and 2.3 for the hydrogen engine), with a carbon reduction of roughly 85 percent (6.8-fold) against grid-power production. Grid parity is not purchasable at any budget. On-site supply is an access and depth product; most current investment targets the wrong term.

Environmental and Economic Implications of Artificial Intelligence Data Centers in the United States

Johanna Bolaños-Zuñiga, Alberto J. Lamadrid

Published 2026-08-10 · arXiv · Credibility S

In this study, we use electricity demand growth, cooling requirements, and backup system operation to evaluate the environmental and economic implications of artificial intelligence data centers in the United States. Our results indicate that impacts are not determined solely by facility design, but by the broader electricity, water, and land-use systems in which these facilities operate. Emissions are primarily dri…

Abstract, interpretation and reference

Abstract

In this study, we use electricity demand growth, cooling requirements, and backup system operation to evaluate the environmental and economic implications of artificial intelligence data centers in the United States. Our results indicate that impacts are not determined solely by facility design, but by the broader electricity, water, and land-use systems in which these facilities operate. Emissions are primarily driven by electricity consumption and therefore depend on marginal generation mixes, transmission constraints, and the spatial and temporal distribution of demand. Analysis further shows that local effects include pressures on water resources, increased noise exposure, and land-use changes, with outcomes varying across regions and infrastructure conditions. The assessment of technological and operational measures shows that improvements in energy efficiency, cooling configurations, and operational strategies can reduce these impacts, although their effectiveness depends on system-level conditions. Evaluation of regulatory and market structures suggests that existing frameworks may not fully account for location- and time-specific externalities. These findings support the need for integrated policy approaches that align data center deployment and operation with electricity system characteristics, water availability, and land-use planning to improve overall environmental and economic performance.

Predictive Failure Detection in Network Hardware Using Thermal Imaging and Deep Learning with Sensor Fusion

Ashly Joseph

Published 2026-08-05 · arXiv · Credibility S

Unplanned network hardware malfunctions can interrupt services and result in expensive downtime in data centers. A deep learning-based predictive maintenance strategy is presented that utilizes thermal imaging and power sensor data to detect early indicators of equipment breakdown in routers, switches, and servers. A simulated dataset was generated comprising annotated thermal pictures and power readings indicative …

Abstract, interpretation and reference

Abstract

Unplanned network hardware malfunctions can interrupt services and result in expensive downtime in data centers. A deep learning-based predictive maintenance strategy is presented that utilizes thermal imaging and power sensor data to detect early indicators of equipment breakdown in routers, switches, and servers. A simulated dataset was generated comprising annotated thermal pictures and power readings indicative of three operating states: Normal, Warning, and Critical. Three ImageNet-pretrained convolutional neural network (CNN) models ResNet-50, InceptionV3, and VGG16 were assessed together with a multi-modal CNN-LSTM fusion model that integrates visual and sensor time-series information. Experiments were performed with and without pre-processing procedures, including region-of-interest (ROI) extraction and normalization. In the absence of pre-processing, CNNs attained moderate accuracy (e.g., ResNet-50 at 52%), but ROI-based pre-processing significantly enhanced performance (ResNet-50 accuracy reaching 91%). The CNN-LSTM model attained the greatest accuracy of 94%, with precision and recall approaching 95%, illustrating the effectiveness of multi-modal fusion. The results validate that domain-specific pre-processing and sensor fusion substantially improve early failure prediction, providing a potential foundation for proactive maintenance of network hardware through non-intrusive monitoring.

A Stackelberg-Bayesian Capacity-Market Game of Carbon Regulation and Second-Life Battery Investment under AI Data-Center Load Growth

Rouzbeh Haghighi, Ali Hassan, Sina Mohammadi, Marcus Chen I Wada, Wencong Su

Published 2026-08-04 · arXiv · Credibility S

Artificial intelligence (AI) data centers are driving rapid electricity load growth across all U.S. ISO/RTO regions, raising both system costs and carbon exposure. This study develops a three-level Stackelberg--Bayesian game in which a regulator (leader) sets carbon penalties and subsidies, a single ISO capacity market clears against an energy balance modeled as a classical generation-expansion problem, and technolo…

Abstract, interpretation and reference

Abstract

Artificial intelligence (AI) data centers are driving rapid electricity load growth across all U.S. ISO/RTO regions, raising both system costs and carbon exposure. This study develops a three-level Stackelberg--Bayesian game in which a regulator (leader) sets carbon penalties and subsidies, a single ISO capacity market clears against an energy balance modeled as a classical generation-expansion problem, and technology-specific investors (followers) decide capacity and operation under incomplete information, yielding a Bayesian Nash equilibrium. The AI impact is captured parsimoniously as an additional load-growth factor on a greenfield-incremental expansion, isolating how much new capacity the growth pulls in and which technology fills it. Within this framework, we consider second-life battery (SLB) storage competing against new/first-life storage for capacity-market revenue. We quantify how a carbon tax, a renewable subsidy, and an SLB subsidy reshape the equilibrium investment mix, carbon emissions, and profit. Different scenarios are compared at the end based on cost-effectiveness and reduced carbon emissions.

From Individual to Shared Ownership: A Coalitional Game Approach to Sustainable Co-investment

Published 2026-07-29 · arXiv · Credibility S

This paper proposes a cooperative game-theoretic framework for sustainable co-investment in shared infrastructure under regulatory incentives. Multiple heterogeneous operators co-invest in a common infrastructure whose production capability evolves over time and is subject to operational variability. A regulator supports the deployment through incentive mechanisms designed to align individual economic investment obj…

Abstract, interpretation and reference

Abstract

This paper proposes a cooperative game-theoretic framework for sustainable co-investment in shared infrastructure under regulatory incentives. Multiple heterogeneous operators co-invest in a common infrastructure whose production capability evolves over time and is subject to operational variability. A regulator supports the deployment through incentive mechanisms designed to align individual economic investment objectives with the coalitional one. We formulate the co-investment problem as a transferable-utility (TU) coalitional game in which the value generated by cooperation depends on heterogeneous operational profiles, dynamic resource availability, investment costs, and regulatory incentive level. We show that the proposed coalitional game can be reformulated as a linear production game (LPG), whose dual prices yield a constructive and stable allocation of the cooperative surplus. Finally, we illustrate the proposed framework through a case study on co-investment among data center operators in shared renewable energy infrastructure, supported by government subsidies promoting renewable energy consumption.

A Configurable Thermal-Dynamic Model for AI Data Center Cooling Load Simulation

Cletus Ngwerume, Lang Tong, Chee-Wooi Ten, Yi Hu

Published 2026-07-31 · arXiv · Credibility S

Cooling demand constitutes a significant and flexible component of AI data center electricity consumption, but time-synchronized measurements are scarce and constant coefficient-of-performance models cannot represent thermal dynamics. This letter proposes a configurable thermal dynamic simulation model for hybrid air- and liquid-cooled data centers. Unlike existing models centered on temperature prediction or equipm…

Abstract, interpretation and reference

Abstract

Cooling demand constitutes a significant and flexible component of AI data center electricity consumption, but time-synchronized measurements are scarce and constant coefficient-of-performance models cannot represent thermal dynamics. This letter proposes a configurable thermal dynamic simulation model for hybrid air- and liquid-cooled data centers. Unlike existing models centered on temperature prediction or equipment-level cooling analysis, the proposed model is designed to generate dynamic cooling electricity profiles for long-duration power system studies. The model is validated using operational telemetry from the Marconi100 supercomputer. Compared with the baseline, the proposed model reduces the mean absolute error from 95.80 to 20.88~kW and the root-mean-square error from 109.79 to 27.27~kW. Evaluation over approximately 520 daily profiles further shows improved reproduction of daily peak demand and intraday variability. The proposed model provides a computationally tractable means of generating physically interpretable cooling load profiles for power system studies.

Real-Time Control of Sustainable Data Centers: A Two-Layer Model Predictive Control Framework with Workload Flexibility and Heat Recovery

Wenyu Liu, Enea Figini, Mario Paolone

Published 2026-08-17 · arXiv · Credibility S

This paper proposes a two-layer model predictive control (MPC) framework for the real-time operation of data centers integrated with on-site photovoltaic generation, battery energy storage, waste heat recovery, and district heating. The upper layer employs scenario-based stochastic optimization to jointly optimize intraday market participation, workload scheduling, and energy management under uncertainty. The lower …

Abstract, interpretation and reference

Abstract

This paper proposes a two-layer model predictive control (MPC) framework for the real-time operation of data centers integrated with on-site photovoltaic generation, battery energy storage, waste heat recovery, and district heating. The upper layer employs scenario-based stochastic optimization to jointly optimize intraday market participation, workload scheduling, and energy management under uncertainty. The lower layer adopts an adaptive tube-based MPC strategy that compensates short-term disturbances while tracking the dispatch references given by the upper layer. The framework further integrates multi-horizon forecasting to support real-time decision making. Microservice-based simulation studies under representative clear-sky and overcast operating conditions demonstrate that the proposed framework accurately tracks dispatch plans despite fast photovoltaic and workload fluctuations. Compared with single-layer control strategies, the adaptive lower-layer controller substantially reduces real-time dispatch deviations and the associated imbalance costs. In addition, the proposed framework naturally adapts to seasonal operating conditions and responds to carbon-aware operating signals, offering a practical approach for economically efficient, sustainable, and grid-supportive operation of future data centers.

Emission-Forecasting-Based Spatial-Temporal Carbon Response: A Multi-Agent Attention-Enhanced Deep Learning Framework

Feiyu Cai, Jing Qiu, Yi Yang, Chenxi Zhang, Xinlei Wang, Baichuan Liu, Junhua Zhao

Published 2026-07-29 · arXiv · Credibility S

As a major contributor to carbon emissions, the decarbonization of power systems has garnered significant societal attention. Nodal carbon intensity (NCI), a critical factor in carbon-oriented demand response, has traditionally been determined through ex-post calculations. However, this ex-post approach introduces latency in low-carbon dispatch. To address this, this paper presents a proactive ex-ante spatial-tempor…

Abstract, interpretation and reference

Abstract

As a major contributor to carbon emissions, the decarbonization of power systems has garnered significant societal attention. Nodal carbon intensity (NCI), a critical factor in carbon-oriented demand response, has traditionally been determined through ex-post calculations. However, this ex-post approach introduces latency in low-carbon dispatch. To address this, this paper presents a proactive ex-ante spatial-temporal carbon response framework. At its core, we develop a novel deep learning-based hierarchical design, enhanced by a dual-stage attention mechanism and a large language model (LLM)-based multi-agent cooperation system, to accurately forecast day-ahead NCI. This design effectively mitigates the impact of renewable energy uncertainty and enhances predictive resilience. On the demand side, the framework proposes a spatial-temporal carbon scheduling model that integrates geographically dispatchable loads (GDLs), including mobile energy storage systems (MESSs) and distributed data centers (DDCs). Leveraging high-accuracy day-ahead NCI predictions, the framework can effectively reduce system emissions by quickly responding to carbon intensity fluctuations. The proposed framework is tested on the modified IEEE 33-bus system. According to the simulation results, the impacts of proposed framework on dispatching latency and emission outcomes are analyzed. The results demonstrate that under a one-hour reduction in carbon scheduling latency, the proposed model and methodology can achieve over 30% emission reduction. This research breaks through the limitations of passive carbon accounting, advancing toward proactive carbon management. It offers an intelligent solution that accelerates the transition to cleaner power systems while directly supporting sustainable production goals.

Inter-Area Oscillation Damping in Data-Center-Integrated Power Systems

Ahmed Alfatlawi, Masoud H. Nazari

Published 2026-07-30 · arXiv · Credibility S

This paper develops explicit dynamic models of a hyperscale data center, including its heating, ventilation, and air conditioning (HVAC) and uninterruptible power supply (UPS) subsystems, and integrates them into a small-signal stability framework to investigate the impact of data center demand response on power system inter-area oscillations. Through eigenvalue analysis and time-domain simulations, the results demo…

Abstract, interpretation and reference

Abstract

This paper develops explicit dynamic models of a hyperscale data center, including its heating, ventilation, and air conditioning (HVAC) and uninterruptible power supply (UPS) subsystems, and integrates them into a small-signal stability framework to investigate the impact of data center demand response on power system inter-area oscillations. Through eigenvalue analysis and time-domain simulations, the results demonstrate that UPS-based demand response can enhance inter-area oscillation damping. In contrast, the HVAC subsystem is shown to be inherently incapable of providing effective oscillation damping due to its limited thermal response bandwidth. A gradient-based optimization algorithm is used to tune the UPS controller gain to maximize the damping ratio of the critical inter-area mode. The effectiveness of the proposed approach is validated using the IEEE 39-bus test system.

AIDC Microgrid Vulnerability Assessment Under Computing-Power Coordinated Attacks

Ze Yu, Hongwei Zhen, Chao Shen, Mingyang Sun

Published 2026-08-11 · arXiv · Credibility S

The rapid growth of large language model (LLM) services is accelerating the expansion of AI data centers (AIDCs), intensifying concerns over power system resource adequacy and rising carbon emissions. The integration of renewable energy provides a pathway toward addressing these pressures, but it also introduces new cross-domain stability challenges to low-carbon AIDCs. For example, variability in renewable generati…

Abstract, interpretation and reference

Abstract

The rapid growth of large language model (LLM) services is accelerating the expansion of AI data centers (AIDCs), intensifying concerns over power system resource adequacy and rising carbon emissions. The integration of renewable energy provides a pathway toward addressing these pressures, but it also introduces new cross-domain stability challenges to low-carbon AIDCs. For example, variability in renewable generation affects reliability on the supply side, whereas fluctuations in AIDC workloads affect reliability on the demand side, jointly creating interconnected stability risks in AIDC microgrids. To address this problem, this paper is the first to explore computing-power coordinated attacks against low-carbon AIDCs. First, we propose an uncertainty-aware AIDC microgrid vulnerability assessment framework to capture two interacting attack surfaces: inverter control parameter tampering attacks, and AI-induced demand manipulation attacks. Then, accounting for renewable-side forecast uncertainty and AIDC-side demand response uncertainty, we introduce confidence-weighted realizations and construct a long-term attack reachable domain analysis. Furthermore, an impedance-based screening method is utilised to map generation and load variations to erosion of stability margin, thereby identifying vulnerable attack time windows and attack vectors. In addition, case studies show that computing-power coordinated attacks induce sustained inverter frequency excursions exceeding 20% of the nominal value and reach instability conditions unattainable by single attacks. The results also demonstrate that the proposed framework can extract sparse, high-confidence vulnerable periods from long-term operating trajectories.

ClusterBench: A Framework for Cluster-Wide Continuous Benchmarking and Regression Testing

Aditya Ujeniya, Jan Eitzinger, Thomas Gruber, Georg Hager, Gerhard Wellein

Published 2026-08-11 · arXiv · Credibility S

Data centers need tooling that validates an entire installation rather than individual nodes, at acceptance and at regular intervals thereafter. This requires dispatching identical benchmarks to every node in a single submission, and therefore cluster-aware scheduling. This paper presents ClusterBench, a framework for cluster-wide continuous benchmarking. It ships with a benchmark collection targeting each component…

Abstract, interpretation and reference

Abstract

Data centers need tooling that validates an entire installation rather than individual nodes, at acceptance and at regular intervals thereafter. This requires dispatching identical benchmarks to every node in a single submission, and therefore cluster-aware scheduling. This paper presents ClusterBench, a framework for cluster-wide continuous benchmarking. It ships with a benchmark collection targeting each component: CPU, GPU, memory, interconnect, and I/O. Because measurements are repeated throughout the cluster's lifetime, ClusterBench collects data across space and time. Comparison against earlier runs detects performance regressions introduced by software changes, such as kernel updates or new library versions. The measurements also form a dataset for research on hardware variability. On the NHR@FAU clusters Helma, Alex, and Fritz, variation within a single component stays within 1%. Variation across specimens reaches 5%, despite nodes identical by specification. Correlating performance with power draw, frequency, and temperature shows that this relationship differs between air- and liquid-cooled nodes.

A Theory of Probabilistic Power Provisioning for Data Centers with Distributed Energy Storage

Can Emre Koksal, Richard A. Barry, Artun Sel

Published 2026-08-13 · arXiv · Credibility S

The growing power demands and variability of AI workloads make electrical power delivery a critical constraint in data-center operation. Distributed energy storage can reduce the power capacity required to support stochastic loads, but its benefits depend fundamentally on the statistics and time scales of demand. This paper develops a probabilistic framework that jointly characterizes provisioned power, energy-stora…

Abstract, interpretation and reference

Abstract

The growing power demands and variability of AI workloads make electrical power delivery a critical constraint in data-center operation. Distributed energy storage can reduce the power capacity required to support stochastic loads, but its benefits depend fundamentally on the statistics and time scales of demand. This paper develops a probabilistic framework that jointly characterizes provisioned power, energy-storage capacity, and the probability of overdraw. We show that storage-assisted provisioning separates into two operating regimes. In the Small Battery Region, overdraw is dominated by short-lived demand excursions and storage provides nearly linear reductions in the required power margin. In the Large Battery Region, overdraw results from sustained demand fluctuations over longer spans of time, and the required margin exhibits diminishing returns with storage. For this regime we introduce effective power, an analogue of effective bandwidth that captures the temporal statistics of the demand and gives an asymptotically tight characterization of the required power. We further quantify how temporal correlation and spatial aggregation affect storage requirements and statistical multiplexing gains, and extend the analysis to loads with multiple demand time scales. Finally, we evaluate the framework using power-demand traces from three production data centers spanning HPC, GPU-training, and cloud-service workloads. Despite their heterogeneous, cyclo-stationary and multi-modal behavior, the measured workloads exhibit the predicted regimes, and a simple four-parameter two-state model captures the dynamics governing their storage-power tradeoffs. The resulting framework provides both a probabilistic foundation and practical dimensioning principles for storage-assisted power provisioning in next-generation AI data centers.

Towards Terabit/$λ$/s Multidimensional Silicon Photonic Engine

Hao Chen, Zengqi Chen, Wu Zhou, Kaihang Lu, Mingyuan Zhang, Yuxiang Yin, Yiou Cui, Chaoran Huang, Pui-In Mak, Yeyu Tong

Published 2026-08-12 · arXiv · Credibility S

Increasing artificial intelligence (AI) workloads drive co-packaged optics (CPO), which integrates optical engines with electronic components. Optical interconnects can extend transmission distances and reduce latency, allowing distributed clusters in AI factories to operate as a unified computational unit. However, escalating data throughput necessitates greater parallelization of light within ultracompact form fac…

Abstract, interpretation and reference

Abstract

Increasing artificial intelligence (AI) workloads drive co-packaged optics (CPO), which integrates optical engines with electronic components. Optical interconnects can extend transmission distances and reduce latency, allowing distributed clusters in AI factories to operate as a unified computational unit. However, escalating data throughput necessitates greater parallelization of light within ultracompact form factors while maintaining stringent energy efficiency and latency constraints. Here, we present a multidimensional silicon photonic engine that achieves a communication capacity exceeding 1.8 terabit/s/lambda/s. By monolithically integrating transceivers, spatial and polarization (de)multiplexers, and optical signal processors on a single chip, we eliminate bulky discrete (de)multiplexers and power-hungry digital signal processing (DSP). In experiments, the photonic engine can be self-configured to identify two, four, or six concurrent spatial and polarization channels per fiber while mitigating dynamic channel crosstalk. Compared with the state-of-art DSP, our approach achieves >5,000-fold reductions in both power consumption and processing latency at a MIMO processing order of six. Furthermore, we demonstrate full-duplex, modulation-format-transparent inter-chip communication over 300-meter fiber. These results represent a paradigm shift for optical engines in future high-performance computing and AI-driven data centers.

InFactPlanner: Planning Sustainable Geo-Distributed LLM Data Centers

Nicoletta Tsiopani, Moysis Symeonides, George Pallis, Marios D. Dikaiakos

Published 2026-08-13 · arXiv · Credibility S

The rapid growth of LLM inference is shifting sustainability concerns from one-time training to continuous serving, where infrastructure decisions shape energy use, carbon emissions, water consumption, and service quality. Yet operators often need to compare deployment alternatives before large-scale infrastructure is built, making direct measurement costly, slow, and sometimes infeasible. We present InFactPlanner, …

Abstract, interpretation and reference

Abstract

The rapid growth of LLM inference is shifting sustainability concerns from one-time training to continuous serving, where infrastructure decisions shape energy use, carbon emissions, water consumption, and service quality. Yet operators often need to compare deployment alternatives before large-scale infrastructure is built, making direct measurement costly, slow, and sometimes infeasible. We present InFactPlanner, a trace-driven decision-support framework for what-if analysis of sustainable AI data center deployment for LLM inference across single and geo-distributed sites. InFactPlanner combines query traces, hardware-model profiles, candidate site configurations, PUE/WUE parameters, renewable generation models, and time-varying grid carbon intensity to estimate power, energy, carbon emissions, water use, latency, and server utilization. The framework abstracts low-level serving effects into configurable hardware-model profiles, enabling rapid comparison of site selection, capacity placement, hardware, model, renewable integration, and routing choices. We validate the energy accounting pipeline by reproducing reference LLM inference energy estimates with less than 10% deviation, evaluate scalability across multiple data centers and server counts, and demonstrate scenario-driven decision analyses for hardware selection, renewable placement, geographic deployment, and carbon-aware routing. Our results show that sustainability-optimal choices can differ from latency-optimal ones, and that the carbon value of deployment depends strongly on the local grid mix.

Planning Waste-to-Energy-Coupled AI Data Centers Through Grade-Matched Cooling and Corridor Screening

Qi He, Chunyu Qu, Wenjie Zuo

Published 2026-07-28 · arXiv · Credibility S

AI data-center growth is increasingly constrained by limited deliverable electricity, interconnection capacity, and cooling demand. This study develops a boundary-consistent screening framework for waste-to-energy (WtE)-coupled AI data-center cooling. It treats cooling as an energy service that can be supplied through grade matching rather than only through electricity-driven mechanical chilling. The framework trans…

Abstract, interpretation and reference

Abstract

AI data-center growth is increasingly constrained by limited deliverable electricity, interconnection capacity, and cooling demand. This study develops a boundary-consistent screening framework for waste-to-energy (WtE)-coupled AI data-center cooling. It treats cooling as an energy service that can be supplied through grade matching rather than only through electricity-driven mechanical chilling. The framework translates plant-side exportable heat into corridor-level planning metrics by accounting for thermal attenuation, absorption conversion, and parasitic electricity for delivery and auxiliaries. In a reference case, a regulated WtE plant processing 1500 t/day of municipal solid waste at 10 MJ/kg provides about 78.1 MWth of exportable heat. At a 20 km corridor, this yields about 53.0 MW of delivered cooling and 8.0 MWe of net avoided cooling electricity after parasitic loads. The coupled system is governed by operating regimes rather than a single efficiency score. Under baseline assumptions, full thermal coverage extends to about 20.9 km, the quality-adjusted criterion remains positive to about 22.9 km, and net electricity relief remains positive to about 44.7 km. For a 1 GW IT campus at 70 percent utilization and a 5 km corridor, net grid relief ranges from about 116.9 to 264.4 MW across scenarios. The required WtE footprint ranges from about 3 to 148 representative plants, or 0.6 to 40 full-load-equivalent plants at a 25 percent displacement target. The framework identifies when WtE-coupled cooling is corridor-feasible, when hybrid operation is required, and when infrastructure scale becomes the binding constraint. It is intended for screening and comparison, not project-specific hydraulic or plant-cycle design.

Techno-Economic Boundary Analysis of Small Modular Reactor Cogeneration for Hyperscale Data Center IT and Cooling Loads

Honglin Li, Buxin She, Jie Zhang

Published 2026-08-11 · arXiv · Credibility S

Hyperscale data centers are adding firm, high-utilization demand faster than grids can serve it, renewing interest in colocating them with small modular reactors. Such a plant could earn revenue in two ways, selling low-carbon power and diverting steam to absorption chillers that serve a cooling load accounting for 20-40% of facility electricity use, but neither revenue stream has been priced across the conditions t…

Abstract, interpretation and reference

Abstract

Hyperscale data centers are adding firm, high-utilization demand faster than grids can serve it, renewing interest in colocating them with small modular reactors. Such a plant could earn revenue in two ways, selling low-carbon power and diverting steam to absorption chillers that serve a cooling load accounting for 20-40% of facility electricity use, but neither revenue stream has been priced across the conditions that must coincide. Here we co-optimize reactor dispatch, steam extraction, absorption cooling and grid exchange hourly for a 200 MW$_\mathrm{e}$ data center in the Electric Reliability Council of Texas (ERCOT) region, across 109 runs spanning capital, market, policy, financing and cooling efficiency. At 2023 mid-range reactor capital, the nuclear configurations cost 49-62% more than grid supply even with the Section 45Y production tax credit. The viable region opens near \$5,000 kW$_\mathrm{e}^{-1}$, and nth-of-a-kind capital makes them 77-89% cheaper in 2023, though between parity and 34% more expensive in the low-price 2024 market. A carbon price of \$53-64 tCO$_2^{-1}$ closes the mid-range gap under hourly export crediting. Absorption cooling is dispatched in response to hourly electricity prices and supplies 38% of annual cooling, at an added cost of \$9.2 million yr$^{-1}$ relative to the reactor-only plant; that gap closes at an installed absorption cost of \$60 kW$_\mathrm{c}^{-1}$ at baseline efficiency and \$570 kW$_\mathrm{c}^{-1}$ on a legacy-efficiency campus, against surveyed commercial prices of \$450-1,200 kW$_\mathrm{c}^{-1}$. Together these results delineate the capital, market and policy conditions under which colocated reactor cogeneration is competitive with grid procurement, and the range over which each condition moves the outcome.

Zero-change foundry compatible silicon photonics MEMS optical switch

Arkadev Roy, Daniel Klawson, Jianheng Luo, Yiyang Zhi, Sirui Tang, Ming Wu

Published 2026-08-04 · arXiv · Credibility S

Large-scale photonic switches are emerging as essential devices for energy-efficient optical interconnect in data centers and AI/ML clusters as a key enabler for high-bandwidth and low-latency connectivity. Combining micro-electro-mechanical (MEMS) based mechanical reconfigurability with silicon photonic integrated circuits can enable a large-scale, low-loss, programmable platform required for large-scale optical ci…

Abstract, interpretation and reference

Abstract

Large-scale photonic switches are emerging as essential devices for energy-efficient optical interconnect in data centers and AI/ML clusters as a key enabler for high-bandwidth and low-latency connectivity. Combining micro-electro-mechanical (MEMS) based mechanical reconfigurability with silicon photonic integrated circuits can enable a large-scale, low-loss, programmable platform required for large-scale optical circuit switches. We demonstrate a broadband silicon photonics MEMS switch with more than 30 dB extinction ratio operating in C-band using a zero-change foundry-compatible process and Back-end-of-Line (BEOL) post-processing. The optical switch element exhibits an insertion loss of less than 1.5 dB with a low static power consumption of approx 20 nW at maximum actuation voltage. Our results illustrate that MEMS-based silicon photonics modulators and phase shifters can be used alongside standard silicon photonics components seamlessly in scenarios where performance in terms of footprint, extinction ratio, broad bandwidth, and low-loss operation is of paramount importance.