智算中心论文观察｜2026-09-10

Current Issue

Volume 2026 · Issue 09-10

按期刊卷期页方式整理本期论文。每条仅使用日报已列出的可追溯公开来源，不新增未经核验事实。

Research Article算电协同

Grid-Mode-Aware Model Predictive Control of Hybrid Energy Storage Systems for AI Data Center Power Smoothing

Xin Chen

Published 2026-09-03 · arXiv · Credibility S

Abstract, interpretation and reference

Abstract

To facilitate the grid-friendly integration of highly variable AI data center loads, this paper proposes a grid-mode-aware model predictive control (G-MPC) framework for managing a hybrid energy storage system (HESS) to smooth grid-side power demand. The framework optimally coordinates a battery energy storage system (BESS) and a supercapacitor (SC) by solving a multi-step optimization problem in a receding-horizon manner. In particular, band-pass filter dynamics are directly embedded in the G-MPC formulation to extract and suppress grid-side power components associated with vulnerable grid oscillatory modes, thus mitigating load-induced grid oscillations. The resulting G-MPC optimization jointly minimizes violations of grid-side power-envelope, ramp-rate, and modal-power requirements and the degradation and power-ramping costs of the BESS and SC, while satisfying power limits, state-of-charge limits, and other operational constraints. To enable real-time implementation, a fix-and-re-optimize algorithm is developed to solve each G-MPC problem efficiently while preventing simultaneous charging and discharging. Extensive simulations demonstrate the effectiveness, flexibility, and computational efficiency of the proposed framework. The results also highlight the importance of explicitly suppressing power components associated with vulnerable grid modes, rather than merely reducing overall load variations, to effectively mitigate grid oscillations.

Full text 中文海报

Research Article算电协同

Spatial LLM Workload Shifting Needs Foresight: Model Commitment for AI Data Center Operation under Power Grid Constraints

Bojun Du, Hongyang Jia, Tonghui Li, Qingchun Hou, Ze Wang, Ershun Du, Ning Zhang

Published 2026-09-09 · arXiv · Credibility S

Abstract, interpretation and reference

Abstract

AI data centers may face power supply shortages during certain periods, requiring operators to shift large language model (LLM) inference workloads spatially to maintain service rates. However, existing workload-shifting methods typically assume that any data center with sufficient computing resources can immediately serve shifted requests, which may lead to infeasible transfers and unserved demand. This letter proposes model commitment (MC), a mixed-integer linear programming framework that jointly schedules model deployment and cross-site request routing under power constraints and electricity-price signals. First, MC formulates the intertemporal coupling introduced by model replica loading. Second, it translates prefill and decode latency requirements into the amount of demand that each replica can serve. Case studies based on real-world data show that MC enables AI data center operators to achieve a 100% service rate under time-varying grid conditions and reduce total operating cost by 29.0%.

Full text 中文海报

Research Article算电协同

Exploiting the Benefits of V2B Application on Peak Shaving of Data Center Loads

Arya Joshi, Hamed Haggi, Chinmay Morankar

Published 2026-08-31 · arXiv · Credibility S

Abstract, interpretation and reference

Abstract

The accelerated growth in data center projects has introduced a demand-driven bottleneck throughout power grids and contributed to a substantial increase in carbon emissions. These concerns are fueling discussions on methods to use existing energy assets to drive operational efficiency. To this end, this paper explores the benefits of Vehicle-to-Building (V2B) applications to support peak shaving of data center cooling loads. Initially, a literature review was conducted considering V2B constraints and optimization methods including SoC limitations, EV participation, tariffs, and building loads. This analysis was then used to develop a conceptual case study of a 10 MW data center in Loudoun County, VA by simulating a temperature-dependent load profile and adjusting the V2B participation of 40 commercial and passenger EVs. Simulation results indicate that, depending on seasonal variations in cooling load demands, strategic deployment of V2B assets between 12-5pm can offset gross cooling loads by 13-36%.

Full text 中文海报

Research Article算电协同

Flexible Training Workloads in Large-Scale AI Data Centers for Transient-Stability Support in Transmission-Constrained Power Systems

Jae-Kyeong Kim

Published 2026-08-31 · arXiv · Credibility S

Abstract, interpretation and reference

Abstract

The rapid expansion of large-scale artificial intelligence (AI) data centers is adding substantial, concentrated, and rapidly varying loads to transmission-constrained power systems. Although such load variations are generally regarded as operational challenges, this paper presents an alternative perspective in which the upward load flexibility of AI data centers could be coordinated for transient-stability support. To this end, this paper proposes training-induced load surge (TILS), a fast demand-side strategy that initiates or resumes flexible AI training workloads after fault clearing to increase active-power demand at electrically effective locations. The resulting load increase allows accelerating generators to supply additional electrical power, thereby reducing the accelerating-power imbalance and limiting the first-swing rotor-angle excursion. The underlying mechanism is first clarified in a single-machine infinite-bus (SMIB) system and then evaluated in the IEEE 39-bus system and a large-scale Korean power system. Results across all three systems demonstrate that TILS can increase the transient-stability-constrained generation limit. Larger responses, earlier activation, and siting at buses with a stronger electrical influence on the critical generators provide greater generation-limit increases. These results suggest that the upward load-response capability of AI data centers can provide complementary transient-stability support when sufficient electrical headroom, flexible workloads, and reliable grid-triggered activation are available.

Full text 中文海报

Research Article算电协同

Hosting Capacity Assessment of Data Centers with Voltage Ride-Through Capability in Power Systems

Pengyu Ren, Wei Sun, Fei Teng

Published 2026-09-02 · arXiv · Credibility S

Abstract, interpretation and reference

Abstract

Large data centers are emerging as concentrated, power-electronic grid loads whose abrupt disconnection or transfer to on-site backup supply during voltage disturbances can remove large demand from the power system, and may create a system-level stability problem. Their interconnection feasibility therefore depends not only on steady-state thermal and voltage limits, but also on whether internal power-conditioning systems can maintain IT service while limiting customer-initiated load reduction. This paper presents a voltage ride-through (VRT)-aware data center and grid co-planning framework that couples transmission-level fault simulation with an internal data center ride-through model. Python-based dynamic simulations generate point-of-interconnection (POI) voltage trajectories under selected network faults, and the resulting waveforms drive an internal model incorporating IT and cooling-load dynamics, DC-link, Uninterruptible Power Supply (UPS) response, and converter apparent power limits. The IEEE 118-bus case study shows that internal VRT capability can become a binding interconnection constraint: steady-state planning alone can overestimate feasible data center capacity, whereas increased UPS converter headroom progressively restores hosting capacity. Under the reduced-order response models studied, the grid-forming mode provides greater ride-through margin than the current-limited grid-following mode under the same network fault conditions. The results further show that VRT constraints can materially change both the total hosting capacity of data centers and its spatial allocation across candidate interconnection buses.

Full text 中文海报

Research Article算电协同

Shift or curtail? How much data-center flexibility is worth depends on the host power grid

Saroj Khanal, Geon Roh, Boyu Yao, Abraham Silverman, Dennice Gayme, Charalambos Konstantinou, Jip Kim, Yury Dvorkin

Published 2026-08-20 · arXiv · Credibility S

Abstract, interpretation and reference

Abstract

Data-center growth risks overbuilding power grid infrastructure and stranding capital. Flexible data-center operation can defer infrastructure investments, but its value depends on the flexibility mechanism and the host power grid characteristics. We classify data-center load as firm, flexible or interruptible, and embed them in capacity expansion applied to market-organized, fossil-heavy PJM and carbon-capped, centrally coordinated Korea. In PJM, the flexibility value is spatial: shifting workloads between zones reduces system cost by 6% in 2028 and 19% in 2038, avoiding 4.4 GW and 8.9 GW of gas and nuclear generation. In Korea, it is temporal: shifting load into midday solar hours makes 0.5 GW of additional solar worth building in 2028 and avoids 1.2 GW of gas and 0.3 GW of batteries in 2038. In both, realistic event-shape limits diminish the value of curtailment. The results show that flexibility procurement and its value are driven by grid characteristics and policy objectives.

Full text 中文海报

Research Article算电协同

Minimizing Grid Interconnection Capacity Requirements for AI Data Centers: A Developer-Side Planning Framework with Onsite Resources and Workload Flexibility

Hassan Zahid Butt, Rida Fatima, Xingpeng Li

Published 2026-08-29 · arXiv · Credibility S

Abstract, interpretation and reference

Abstract

Securing grid interconnection capacity has become a bottleneck for AI data center projects and can take longer than constructing the facilities themselves. This mismatch can delay deployment for years, making early interconnection planning essential. This paper develops ICP-AI, an interconnection capacity planning framework from a data center developer's perspective. The framework minimizes grid import capacity under a prescribed onsite investment budget while jointly sizing photovoltaic (PV) and battery energy storage system (BESS) resources and scheduling deadline constrained workload flexibility. A secondary refinement fixes the minimum grid capacity and selects the minimum-investment PV-BESS portfolio among solutions that achieve that capacity. The framework is evaluated using monthly composite stress profiles across varying temporal assumptions, load shapes, flexible load fractions, and deferral windows. Results show that interconnection capacity reduction depends strongly on the planning environment: at a $100M budget, it is about 6% for the high load factor baseline, exceeds 10% under monthly average solar availability, and reaches 13.3% for a more diurnal load. At a $10M budget, 5% flexible load with a 1 h workload deferral window reduces BESS capacity from 15.30 to 4.87 MWh while increasing capacity reduction from 4.43% to 4.84%. To test sensitivity to temporal compression, the model is also solved over the full 8,760 h chronology, which preserves the main capacity and flexibility trends. Overall, ICP-AI quantifies the interconnection capacity and infrastructure substitution value of workload flexibility, providing an investment-interconnection frontier to support capital allocation and early project planning in constrained grid environments.

Full text 中文海报

Research Article算电协同

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

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.

Full text 中文海报

Research ArticleAI 运维优化

Operations, Maintenance, and Industrial Scaling of MW-Class Orbital Data Centers

Slava G. Turyshev

Published 2026-08-27 · arXiv · Credibility S

Abstract, interpretation and reference

Abstract

Megawatt-class orbital data centers require continuous maintenance, replacement, inventory, and service capacity in addition to spacecraft power/thermal systems. We formulate an analytical lifecycle framework for permanent/transient failures, modular orbital replacement units, robotic servicing, spare inventory, scheduled technology refresh, correlated faults, cybersecurity, optional human support. The model combines nonhomogeneous component hazards, capacity-weighted availability, multiclass robotic-service capacity, Poisson base-stock inventory, replacement-flow accounting, human-support break-even relations. For a 1 MW cluster with 10 active 100 kW nodes, 1 reserve node, ~200 5 kW compute cartridges, low, nominal, high deployed-mass allocations span ~50-75 kg/kW. Assumptions yield 70.2 random or life-limited interventions and 323-349 planned refresh operations/(MW-year), for a total of 393-419 standardized operations/(MW-year). Analysis gives a first-generation logistics of 5.3-9.0 t/(MW-year), with a nominal case of ~ 6.6 t/(MW year), 560-700 productive robot-hors/(MW-year). Planned refresh exceeds random replacement under the stated component populations, hazards, 3-15-year intervals. At ~400 standardized operations/(MW-year), the post-internal-recovery exception probability is <$10^{-3}$, with an objective near $10^{-4}$ at large scale; terminal non-recovery $p_U$ requires a smaller mission-level allocation. The target catastrophic-loss hazard for a 100 kW node is 0.01-0.03 1/yr. Parametric workload and cost cases place contingency visits at 10s of MWs, periodic campaigns at 10-100s of MWs, dedicated personnel at several 100 MWs to GWs. The reference first deployment is uncrewed, autonomously fault-managed, robotically maintainable, supported by specific inventory based on a 6-month replenishment horizon, compatible with later human access without permanent habitation.

Full text 中文海报

Research Article算电协同

InFactPlanner: Planning Sustainable Geo-Distributed LLM Data Centers

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

Published 2026-08-13 · arXiv · Credibility S

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.

Full text 中文海报

Research ArticleAI 运维优化

Quantifying AI data center nitrogen oxide (NO$_x$) emissions from space

Kevin D. Gauld, Daniel J. Varon, Nicholas Balasus, Daniel H. Cusworth

Published 2026-08-23 · arXiv · Credibility S

Abstract, interpretation and reference

Abstract

AI data center power demand is spurring rapid deployment of on- and near-site natural gas turbines. Nitrogen oxide (NO$_x$) pollution from this equipment is a growing concern but has not previously been quantified with atmospheric observations. Here we demonstrate space-based detection and quantification of NO$_x$ emissions from the SpaceXAI Colossus 2 power plant in Southaven, Mississippi. Using observations from the geostationary TEMPO satellite instrument, we detect a strong increase in local mean NO$_2$ column concentrations after the plant began operations in late 2025. We then use TEMPO to estimate two-week-average NO$_x$ source rates from August 2025 to mid-August 2026, calibrating against continuous emission monitoring system (CEMS) data from US power plants. TEMPO first detected NO$_x$ emissions in December 2025 at 460$\pm$180 kg h$^{-1}$. We find that emissions increased through August 2026, averaging 730$\pm$185 kg h$^{-1}$ after February 2026, roughly 16 times higher than expected from the facility's March 2026 permit for 41 turbines operating under best available control technology (BACT) requirements ($\sim$47 kg h$^{-1}$). Emissions at the expected level would be undetectable by our TEMPO analysis.

Full text 中文海报

Research Article余热回收

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

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.

Full text 中文海报

Research Article算电协同

AIDC Microgrid Vulnerability Assessment Under Computing-Power Coordinated Attacks

Ze Yu, Hongwei Zhen, Chao Shen, Mingyang Sun

Published 2026-08-11 · arXiv · Credibility S

Abstract, interpretation and reference

Abstract

The rapid growth of large language model (LLM) services is expanding AI data centers (AIDCs), increasing electricity demand and associated carbon emissions. Renewable energy integration can mitigate these impacts but also strengthens the coupling between AIDC loads and inverter-interfaced generation, creating cross-domain cyber-physical vulnerabilities. Specifically, adversarial AI requests alter AIDC power demand, whereas inverter control tampering modifies source-side dynamics, and their combined impact on system stability varies with generation forecast and demand response uncertainties. To this end, we propose an uncertainty-aware AIDC microgrid vulnerability assessment framework under computing-power coordinated attacks. First, the framework maps adversarial AI requests to AIDC power variations and represents uncertainties in attack-induced demand responses and photovoltaic (PV) forecasts through confidence-weighted realizations. Then, impedance based stability analysis combines these realizations with bounded inverter parameter tampering to construct attack reachable domains and identify critical attack time windows. Furthermore, a separate criterion identifies fixed coordinated attack vectors that retain destabilizing capability throughout each selected window. Case studies demonstrate that, unlike either attack component applied alone, coordinated attacks within identified critical windows induce sustained inverter frequency oscillations with peak absolute deviations exceeding 20% of nominal frequency, whereas the evaluated out-of-window response remains bounded. The proposed method further identifies critical attack windows and the associated coordinated attack vectors.

Full text 中文海报

Research Article芯片与算力

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

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.

Full text 中文海报

Research Article算电协同

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

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.

Full text 中文海报

Research Article算电协同

Real-Time Edge-based Detection of Correlated AI Data-Center Load Episodes

Chandan Chaudhary, Abanish Tiwari, Yansong Pei, Mohammed Ben-Idris, Joydeep Mitra

Published 2026-08-24 · arXiv · Credibility S

Abstract, interpretation and reference

Abstract

Artificial-intelligence data centers running bulk-synchronous training can impose sub-second power swings. When several facilities synchronize their training cycles, these load variations become spatially correlated and amplify the aggregate disturbance on the grid. A grid operator without access to data-center telemetry must infer this correlation from electrical measurements alone. However, the required observation time and the feasibility of detection on substation-deployable hardware remain uncharacterized. This paper develops a correlation-based detection method to classify the multi-facility operating regime from cross-facility power measurements. Analytical derivations and experimental validation show that the resulting detection confidence increases with the observation-window length at a rate governed by the load correlation time. The method is demonstrated in a real-time hardware-in-the-loop testbed, where load setpoints generated from a validated semi-Markov data-center load model are applied to an electromagnetic-transient grid simulation on a Real-Time Digital Simulator. A compact classifier built on pairwise power correlations runs on an edge device in this loop and determines whether the data-center load variations are independent or spatially correlated. The cross-facility correlation separates the independent and correlated cases across independent realizations. The held-out detection accuracy improves with the observation window, consistent with the predicted relation. A raw-waveform network fails to generalize, supporting pairwise correlation as the discriminative signal. The detector executes in real time on commodity edge hardware. A closed-loop demonstration against the running simulator tracks a regime change within one observation window.

Full text 中文海报

Research Article算电协同

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

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.

Full text 中文海报

Research Article芯片与算力

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

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.

Full text 中文海报

Research Article热管理与液冷

Generalizing Thermal Transport in High-Contrast Metamaterials through Interfacial Fresnel Reflection

Seung Hyeon Ham, Yu Min Kim, In Hyeok Choi, Jeong Woo Han

Published 2026-08-26 · arXiv · Credibility S

Abstract, interpretation and reference

Abstract

The rapid growth of generative AI has intensified the need for efficient heat dissipation in large-scale data centers. To control heat flow, thermal metamaterials with layered structures have been widely used, which impart the anisotropic properties of thermal conductivities. However, the conventional effective medium approximation (EMA) often fails to provide accurate predictions in systems with a high thermal conductivity contrast between adjacent layers embedded in a background medium. Here, we generalize the EMA by introducing two corrective coefficients that extend its validity to regimes where the conventional EMA was previously inapplicable, i.e., high-contrast thermal metamaterials with the background medium. Notably, one of these coefficients that we proposed has the same mathematical form as the Fresnel reflection coefficient in optics. This allows us to interpret the "reflection-like" behavior of heat flow as it penetrates adjacent layers with high thermal contrast. Our findings suggest that heat diffusion, traditionally viewed as a purely dissipative process, can be understood intuitively through the framework of ray optics.

Full text 中文海报

Research Article芯片与算力

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

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.

Full text 中文海报

Research Article芯片与算力

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

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.

Full text 中文海报

智算中心论文专站

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract