CS Paper Radar

Data center cooling choices shift water impacts across the grid: An integrated water-energy model for sustainable data center development

Garrett Alston, Nancy Love, Rabab Haider

eess.SY cs.CYcs.DC ·submitted 2026-09-21 21:48 UTC · v1
code/data

Abstract

Data centers are being developed at an unprecedented pace, yet their energy and water impacts, and the spatial and temporal distribution of these impacts, remain poorly characterized. Data centers consume water for cooling (direct) and through electricity generation (indirect). Decisions on siting and cooling technology result in water-energy trade-offs that extend impacts beyond the facility's location. Existing assessment frameworks rely on facility efficiency metrics and average grid water intensity factors, suppressing the temporal impacts of data center load and generation availability. They also attribute indirect consumption to the facility's location rather than to the generators (and corresponding hydrologic regions) that respond to the added load, misattributing spatial impacts. To close this gap, we develop a computational model of the data center-energy-water nexus that links facility cooling and electricity demand with hourly economic dispatch, generator-level water consumption, and monthly subbasin depletion. Built on open-source data, the model resolves where and when water is consumed, and where this consumption compounds existing water risk or creates new risk. Using the model, we study different cooling configurations and proposed developments in the state of Michigan. Air-cooled data centers halve total water consumption relative to evaporative cooling, but increase electricity demand and raise indirect water consumption by one-third, shifting the water footprint from the facility to generators. Mapping these changes to subbasins reveals depletion increases beyond the data center sites, in regions that facility-level reporting may overlook. These results show that data center water and energy impacts cannot be assessed in isolation, motivating the need for integrated modeling to inform siting, design, and reporting practices.

Comment
12 pages; 8 figures; 6 tables; appendix

Closest papers

Nearest neighbours by meaning across the whole index.

GreenDirector: carbon- and water-aware workload placement for sustainable computing

cs.DC ·2026-09-11 ·Jaime Iglesias Blanco, Ignacio Heredia, María Castrillo, Andrei Tsaregorodtsev, Mazen Ezzeddine +1 more
similarity 0.74
Abstract

The rapid growth of data center electricity demand, accelerated by AI, makes carbon-only accounting an incomplete measure of computing's environmental impact: low-carbon electricity mixes are often water-intensive, and the resulting harm depends on local, seasonal scarcity rather than on the volume of water consumed. We propose the Environmental Score (ES), a unified, dimensionless index in $[0, 100]$ that jointly captures the carbon footprint and the spatial-temporal, scarcity-weighted water impact of the electricity a workload consumes. It combines real-time, cross-border electricity flow tracing with monthly AWARE2.0 water-scarcity characterization factors, weighting global greenhouse-gas emissions together with the local, seasonal severity of water stress. Building on it, we define the Green Score (GS), a scheduling metric proportional to the useful computational work delivered per unit of real environmental impact, which also accounts for data center power and hardware efficiency. We add both metrics as a green-affinity feature to the GreenDirector schedulers of two production federated infrastructures, the AI4EOSC scientific cloud and the DIRAC workload management system. In AI4EOSC, a cluster-filling experiment over four pan-European providers shows that greener sites are filled first without degrading scheduling latency or end-user experience. In DIRAC, a trace-driven simulation of 133,631 jobs and a preliminary production deployment for the KM3NeT community reduce carbon emissions and improve carbon efficiency by about 40\%, while making the carbon-water trade-off explicit when the lowest-carbon site also carries higher water stress. The results show that hydrological stress can be dynamically weighted into workload placement in live, multi-tenant systems

0.19

AI Surge Accelerates the Transition to Liquid Cooling: Rethinking Energy and Water Use in Data Center Cooling

cs.CY ·2026-11-18 ·Yingbo Zhang, Eric Masanet, Shengwei Wang
similarity 0.72
Abstract

The power density of AI workloads is rapidly driving shifts to liquid cooling. This study rethinks and assesses the global impact on energy and water demand of this cooling transformation. Results suggest that liquid cooling can reduce cooling energy use by 25-56% relative to air cooling, offering substantial energy savings for GW-scale AI data centers. Best-practice PUE values could be as low as 1.07-1.08 for today's systems and 1.03-1.04 for emerging designs, and WUE values range from 0.38 to 2.63 L/kWh, depending on temperature conditions. We identify global free-cooling rack inlet temperature thresholds of 33 $^\circ$C for evaporative systems and 49 $^\circ$C for dry systems, demonstrating the practicality of 100% free cooling in many locations. The model finds that minimum PUEs do not necessarily coincide with minimum total energy use, because warmer coolant raises server power. Results quantify the large sustainability potential of liquid cooling and provide actionable targets for data center design and policies.

0.20

Artificial Intelligence for Energy Optimization in Data Centers

cs.AI cs.LG ·2026-09-03 ·Mohammed Basharath Ullah, Summaiya Unnisa Begum, Mohammed Nadeem Ullah
code/data similarity 0.68
Abstract · 11 pages, 6 figures, 7 tables

Data centers are increasingly optimized by artificial intelligence and, at the same time, increasingly loaded by it. The literature treats these as two unrelated problems: control studies model workload as an exogenous arrival process, while sustainability studies model infrastructure as a fixed multiplier. We screen roughly 194 papers retrieved through a documented protocol, code 63 of them, and report what the coding shows. Of 28 primary control-oriented studies, 18 are validated in simulation alone and 5 reach physical hardware or a production facility; none account for water withdrawal, and none account for embodied carbon. Reported savings intervals across four technique families overlap almost completely, which means the field cannot presently rank its own methods. Ten recurring gaps are scored for consequence and tractability, and we set out CLEAR-DC, a framework coupling a control-policy branch to a workload-demand branch through an explicit elasticity term, reads out net rather than direct benefit, and emits a schema-conformant record covering energy, carbon, water, embodied share and validation venue. The framework is an architectural and methodological proposal, not a trained system; the contribution we defend empirically is the corpus analysis and the reporting schema derived from it. Coding sheet, derived statistics and all result artifacts: https://github.com/Kimalice/AI-for-Energy-Optimization-in-Data-Centers-Closing-the-Optimizer-Load-Loop

0.45

Modeling Liquid-Cooled Data Centers in EnergyPlus

eess.SY cs.CEcs.DCcs.LG ·2026-10-23 ·Yun-Dam Ko, Dipashreya A. Sur, Cade Napier, Carlos D. Díaz-Marín, Rishee K. Jain
code/data similarity 0.68
Abstract · 14 pages, 7 figures, 6 tables. Code: https://github.com/Urban-Informatics-Lab/energyplus-liquid-cooling

As AI workloads accelerate data center transition toward high-density racks with direct-to-chip (DTC) liquid cooling, existing simulation tools remain limited in evaluating whole-building and multi-year efficiency. To bridge this gap, we develop a native EnergyPlus object, ElectricEquipment:ITE:LiquidCooled, that models DTC liquid cooling for chilled-water and warm-water architectures as plant-loop components. It couples facility-water loops, rack-side secondary loops, coolant distribution units, and chip-junction temperatures within a single solver. Validation against an independent Modelica implementation across 21 operating conditions, including two full-year runs compared at one-minute resolution, yields loop-temperature RMSE at or below 0.021 K, well within the 0.1 K acceptance threshold. Our case study demonstrates that warm-water liquid cooling reduces annual PUE from 1.43 to 1.25, eliminates on-site water use, and cuts total facility energy by 37.9% relative to an air-cooled baseline. By embedding liquid-cooling logic into EnergyPlus, our work provides whole-building, long-term energy modeling for high-density AI infrastructure within an open-source framework.

0.31

Job-Level Workload Allocation for Data Centers Under Flexible Interconnections

eess.SY ·2026-11-14 ·Tanmay Mishra, Arash Omidi, Mads R. Almassalkhi
hot direction similarity 0.66
Abstract · Submitted to IEEE PES General Meeting 2027

Flexible interconnections allow data centers to connect to the grid sooner without upgrading the grid infrastructure, if they accept limits on their grid power during periods of grid stress. When a limit is reached, the data center must slow some of its computing jobs. This choice sets the cost of flexibility, because the work done by a job is worth far more than the energy it uses. This paper models how jobs respond when their power is reduced, and proposes a real-time method that shares the available power among jobs according to their value. It also gives grid planners a simple measure of how much power a data center can reduce, and for how long. A one-year case study of a 100-MW data center shows that the proposed method lowers the cost of meeting grid limits, and that on-site generation and storage lower it further.

0.24

Speed-to-Power Dominates Energy Costs for AI Data Centers: Technoeconomic Analysis of Powering AI in the Continental United States

cs.AI eess.SY ·2026-11-12 ·Alex C. Newkirk, Daniel L. Gerber, Erica R. H. Fuchs, Alex Hubbard, M. Granger Morgan +3 more
code/datahot direction similarity 0.66
Abstract · 80 pages incl. SI

The rapid growth of artificial intelligence is driving demand for new data centers at large scale and on short deployment timelines. To power them, firms connect to the commercial grid or contract generation directly, but interconnection and gas-turbine delays now incentivize on-site generation. We compare grid-connected and off-grid data center energy infrastructure (AC- and DC-coupled solar-plus-storage, islanded natural gas, and grid interconnection) across the continental U.S. Using an open-source technoeconomic framework built on empirically calibrated AI training loads, climate-driven cooling, stage-by-stage power conversion, and physics-based battery degradation, we model designs providing equivalent energy services. Our central contribution is a speed-adjusted cost of energy that values the compute unlocked by faster construction, making options with different deployment timelines directly comparable on one physically consistent basis. We find DC coupling reduces lifetime energy costs by 18% ($0.09/ kWh) on average versus AC counterparts by eliminating inverters and improving efficiency. Though rarely cheapest on unadjusted energy cost, DC-coupled microgrids' faster deployment makes them the lowest-cost modeled option in 85% of U.S. locations once foregone compute is valued, or 55% under accelerated natural gas timelines. For AI infrastructure, deployment speed rather than unit energy cost determines the economically preferred path to power.

0.31

Quantifying the hosting capacity value of workload flexibility for AI data centers: A two-stage electricity-computation co-optimization framework

eess.SY ·2026-11-18 ·Jiakai Wu, Wei Sun, Pengyu Ren, Yifan Wang
hot direction similarity 0.65
Abstract

Grid connection constraints increasingly limit the expansion of AI data centers, yet the additional hosting capacity enabled by computational workload flexibility remains insufficiently quantified. This paper develops a two-stage electricity--computation co-optimization framework to quantify the individual and coordinated hosting capacity values of spatial and temporal workload flexibility. Data center capacities are determined endogenously, with computational task arrivals scaling with admitted capacity. Spatial routing of eligible interactive requests and temporal deferral of batch workloads are linked to server operation and facility power demand, subject to workload completion requirements and transmission constraints. The first stage maximizes aggregate hosting capacity; the second fixes the resulting site capacities and optimizes annual workload scheduling and generation dispatch to minimize generation and carbon costs. An 8760-hour case study with three data centers on a modified IEEE 118-bus system shows that temporal and spatial flexibility individually increase hosting capacity by 21.26\% and 5.37\%, respectively. Their coordinated use raises hosting capacity from 4635.1~MW to 6231.5~MW, a 34.44\% increase without network reinforcement. The coordinated gain exceeds the sum of the individual gains by 361.9~MW, demonstrating complementarity between spatial migration and temporal deferral. By redistributing computational demand across locations and operating hours, these mechanisms enable fuller utilization of existing transmission capacity. The findings establish workload flexibility as a quantifiable planning resource for accommodating additional data center capacity within existing transmission networks.

0.28

Moving bits instead of watts: Accelerating global AI data-centre deployment at the energy-digital-water-land nexus

eess.SY ·2026-10-26 ·Jiawei Qu, Xiaoying Shi, Kai Hou, Hongjie Jia, Jinyue Yan
similarity 0.65
Abstract

Artificial intelligence is driving an unprecedented expansion of data centres. However, this growth is increasingly stalled by grid transmission constraints that trap new data centres in years-long connection queues. Unlike electricity, computation can be moved across the globe through fibre-optic networks, but whether this spatial transfer can bypass constrained grids and accelerate global AI data-centre deployment remains unknown. Here we develop a global framework that places future data-centre capacity in energy-abundant regions and routes computational workloads through fibre-optic networks rather than transmitting electricity. We evaluate the framework through more than 11 million hourly operation simulations at subnational resolution, integrating constraints on energy systems, water, digital connectivity, land and financing risk. Results show that across multiple data-centre development and climate-policy scenarios, leveraging global data transfer allows new data centres to be sited where these conditions are favourable, lowering construction costs by 44-47% and carbon emissions by 38%. By treating the global fibre-optic network as an energy-system asset, it avoids 34-67 GW of high-voltage transmission capacity, comparable in magnitude to Italy's national peak electricity demand. Crucially, this methodology could bring more than 95% of connection-delayed data centres online in 2030, with Canada, the Nordic countries and the Arabian Peninsula emerging as preferred locations for data-centre development. Our findings establish computational mobility as an alternative to energy-system expansion, informing the coordinated global development of energy and artificial-intelligence infrastructure.

0.10