Generative Design for Direct-to-Chip Liquid Cooling for Data Centers

• URL: http://arxiv.org/abs/2604.10941v1
• Date: Mon, 13 Apr 2026 03:20:42 GMT
• Title: Generative Design for Direct-to-Chip Liquid Cooling for Data Centers
• Abstract summary: AI workloads are driving up data center power densities, increasing the need for advanced thermal management.<n>This work presents a generative design framework for cooling channel geometries for the NVIDIA GB200 Grace Blackwell Superchip.<n>A physics-based finite-difference thermal model provides rapid steady-state temperature predictions and supplies spatial thermal feedback.
• Score: 5.435523531334183
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Rapid growth in artificial intelligence (AI) workloads is driving up data center power densities, increasing the need for advanced thermal management. Direct-to-chip liquid cooling can remove heat efficiently at the source, but many cold plate channel layouts remain heuristic and are not optimized for the strongly non-uniform temperature distribution of modern heterogeneous packages. This work presents a generative design framework for synthesizing cooling channel geometries for the NVIDIA GB200 Grace Blackwell Superchip. A physics-based finite-difference thermal model provides rapid steady-state temperature predictions and supplies spatial thermal feedback to a constrained reaction-diffusion process that generates novel channel topologies while enforcing inlet/outlet and component constraints. By iterating channel generation and thermal evaluation in a closed loop, the method naturally redistributes cooling capacity toward high-power regions and suppresses hot-spot formation. Compared with a baseline parallel channel design, the resulting channels achieve more than a 5 degree Celsius reduction in average temperature and over 35 degree Celsius reduction in maximum temperature. Overall, the results demonstrate that coupling generative algorithms with lightweight physics-based modeling can significantly enhance direct-to-chip liquid cooling performance, supporting more sustainable scaling of AI computing.

Related papers

• Virtual Temperature Sensors in Power Transformers Using Neural Ordinary Differential Equations [42.50242605797505]
  This paper develops a physics-aware Neural Ordinary Differential Equation framework for forecasting transformer thermal behavior.<n>A key contribution is the integration of simplified heat-transfer equations directly into the Neural ODE formulation.<n>The framework provides a standardized, physics-aware, and robust forecasting approach for heterogeneous transformer units.
  arXiv  Detail & Related papers  (2026-08-13T13:59:32Z)
• Physics-Informed Machine Learning for Pouch Cell Temperature Estimation [5.435523531334183]
  This paper presents a physics-informed machine learning (PIML) framework for the efficient estimation of steady-state temperature profiles.<n>The PIML approach integrates the governing heat transfer equations directly into the neural network's loss function, enabling high-fidelity predictions.<n>Results demonstrate that the PIML model converges more rapidly and achieves markedly higher accuracy, with a 49.1% reduction in mean squared error over the data-driven model.
  arXiv  Detail & Related papers  (2026-04-16T02:58:10Z)
• 2D-ThermAl: Physics-Informed Framework for Thermal Analysis of Circuits using Generative AI [9.414651358362388]
  'ThermAl' is a physics-informed generative AI framework which effectively identifies heat sources and estimates full-chip transient and steady-state thermal distributions.<n>Our model is trained on an extensive dataset of heat dissipation maps, ranging from simple logic gates (e.g., inverters, NAND, XOR) to complex designs, generated via COMSOL.<n> Experimental results demonstrate that ThermAl delivers precise temperature mappings for large circuits, with a root mean squared error (RMSE) of only 0.71C, and outperforms conventional FEM tools by running up to 200 times faster.
  arXiv  Detail & Related papers  (2025-12-01T00:45:26Z)
• HeatGen: A Guided Diffusion Framework for Multiphysics Heat Sink Design Optimization [0.0]
  Geometries are represented using boundary representations of multiple fins.<n>We train a denoising diffusion probabilistic model to generate heat sinks with characteristics consistent with those observed in the data.<n>We train two different residual neural networks to predict the pressure drop and surface temperature for each geometry.
  arXiv  Detail & Related papers  (2025-11-12T08:07:46Z)
• Cryogenic Control and Readout Integrated Circuits for Solid-State Quantum Computing [44.99833362998488]
  cryogenic integrated circuits (ICs) have emerged as potential alternatives to room-temperature electronics. operating at cryogenic temperatures can suppress electronic noise and improve qubit control fidelity. For CMOS ICs specifically, circuit design uncertainties arise due to a lack of reliable models for cryogenic field effect transistors.
  arXiv  Detail & Related papers  (2024-10-21T11:15:45Z)
• Chemical Reaction Neural Networks for Fitting Accelerating Rate Calorimetry Data [37.69303106863453]
  Chemical Reaction Neural Networks (CRNNs) are trained to fit the kinetic parameters of N-equation Arrhenius ODEs to ARC data obtained from a Molicel 21700 P45B. The flexibility of the method is demonstrated by experimenting with two-equation and four-equation models.
  arXiv  Detail & Related papers  (2024-08-21T20:39:41Z)
• Building Vision Models upon Heat Conduction [66.1594989193046]
  This study introduces the Heat Conduction Operator (HCO) built upon the physical heat conduction principle.<n>HCO conceptualizes image patches as heat sources and models their correlations through adaptive thermal energy diffusion.<n> vHeat achieves up to a 3x throughput, 80% less GPU memory allocation, and 35% fewer computational FLOPs compared to the Swin-Transformer.
  arXiv  Detail & Related papers  (2024-05-26T12:58:04Z)
• Machine Learning-Assisted Thermoelectric Cooling for On-Demand Multi-Hotspot Thermal Management [12.515874333424929]
  Thermoelectric coolers (TECs) offer a promising solution for direct cooling of local hotspots and active thermal management in advanced electronic systems. TECs present significant trade-offs among spatial cooling, heating and power consumption. We present a novel machine learning-assisted optimization algorithm for thermoelectric coolers that can achieve global optimal temperature.
  arXiv  Detail & Related papers  (2024-04-20T18:35:45Z)
• TCI-Former: Thermal Conduction-Inspired Transformer for Infrared Small Target Detection [58.00308680221481]
  Infrared small target detection (ISTD) is critical to national security and has been extensively applied in military areas. Most ISTD networks focus on designing feature extraction blocks or feature fusion modules, but rarely describe the ISTD process from the feature map evolution perspective. We propose Thermal Conduction-Inspired Transformer (TCI-Former) based on the theoretical principles of thermal conduction.
  arXiv  Detail & Related papers  (2024-02-03T05:51:22Z)
• Deep encoder-decoder hierarchical convolutional neural networks for conjugate heat transfer surrogate modeling [0.0]
  Conjugate heat transfer (CHT) analyses are vital for the design of many energy systems.<n>High-fidelity CHT numerical simulations are computationally intensive.<n>We develop a modular deep encoder-decoder hierarchical (DeepEDH) convolutional neural network for CHT analyses.
  arXiv  Detail & Related papers  (2023-11-24T21:45:11Z)
• Machine learning based surrogate models for microchannel heat sink optimization [0.0]
  In this paper, microchannel designs with secondary channels and with ribs are investigated using computational fluid dynamics. A workflow that combines Latin hypercube sampling, machine learning-based surrogate modeling and multi-objective optimization is proposed.
  arXiv  Detail & Related papers  (2022-08-20T13:49:11Z)
• A scalable helium gas cooling system for trapped-ion applications [51.715517570634994]
  A modular cooling system is presented for use with multiple ion-trapping experiments simultaneously. The cooling system is expected to deliver a net cooling power of 111 W at 70 K to up to four experiments.
  arXiv  Detail & Related papers  (2021-06-14T16:37:54Z)
• Uhlmann Fidelity and Fidelity Susceptibility for Integrable Spin Chains at Finite Temperature: Exact Results [68.8204255655161]
  We show that the proper inclusion of the odd parity subspace leads to the enhancement of maximal fidelity susceptibility in the intermediate range of temperatures. The correct low-temperature behavior is captured by an approximation involving the two lowest many-body energy eigenstates.
  arXiv  Detail & Related papers  (2021-05-11T14:08:02Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.