Search NASA⌕ Search

SEARCH · Search NASA

Results for “large-scale computing”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 181 records · Page 10

Foundation models for atomistic simulation of chemistry and materials

Conventional computational methods for modeling chemical and materials systems are limited by system size and timescale, forcing a trade-off between quantum-mechanical accuracy and the sampling needed for realistic observables. Large language and vision foundation models — pre-trained on massive datasets using transformer architectures — have revolutionized many fields. It is thus interesting to ask whether a foundation model — subject to suitable data, parameter scaling and training — could enable learned simulations of chemistry and materials. Here, in this study, we review the field of machine-learned interatomic potentials (MLIPs) and posit that scaling up large and diverse chemical and materials datasets and highly expressive architectures using advanced training strategies should result in models that are: more efficient, transferable, robust to out-of-distribution scenarios, and easier to fine-tune to a variety of downstream physical observables than models trained from scratch on small datasets corresponding to specific, targeted atomistic simulation tasks. We provide specific criteria for creating such large-scale MLIP foundation models, coordinated strategies for their development, evaluation and deployment, and highlight potential emergent capabilities that could transform predictive simulations in chemistry and materials science and accelerate discovery across multiple technological domains.

Yuan, Eric C.-Y. [University of California, Berkel↗

Intermediate time sub-diffusion and stress relaxation in ring polymer melts

The slow dynamics of non-concatenated ring melts remains a frontier problem in polymer science with implications for many soft material environments including cellular biophysics. Here, in this work, we report large-scale simulations of model ring melts that analyze the monomer and center-of-mass (CM) mean square displacements (MSD) and stress relaxation function on intermediate time and length scales. The degree of dynamical slowing down is characterized by the maximally sub-diffusive fractional time scaling exponents. The data span an exceptionally wide range of ring degrees of polymerization and stiffnesses and are not successfully organized based on the classic measure linear chain entanglement, N/N e . Rather, we find that the crossover degree of polymerization, N D , based on ring macromolecular caging that successfully allows master curves to be constructed for the long-time CM self-diffusion constant also collapses these temporal dynamic scaling exponents. Different properties display different exponents and exhibit one or two regimes of linear variation with the logarithm of N D / N . A distinct crossover of the CM-MSD and stress relaxation exponents emerges at sufficiently large N or stiffness that is not found for the monomer MSD, indicating a novel form of dynamic decoupling. This crossover aligns with the predicted critical degree of polymerization for transitioning from a weak to strong caging regime, indicative of activated transport. The latter may reflect the emergence of an intermolecular collective contribution to stress in analogy with dense soft colloidal matter. Suggestions are made for future theoretical work to address the rich patterns of behavior discovered.

Anomalous diffusion↗

Sensitivity of a closed dielectric haloscope to axion dark matter

We present a method to determine the sensitivity of a closed dielectric haloscope to axiondark matter. Dielectric haloscopes aim to probe the theoretically well-motivated axion mass rangeof ∼ 26 µeV to ∼ 500 µeV by utilizing a stack of dielectric disksand a mirror to enhance the axion-photon conversion within an external magnetic field. Theirconversion volume is nearly axion-mass independent, thereby favoring large-scale designs toincrease sensitivity. The large volume causes simulations to be computationally expensive andtime-consuming. This paper presents a simple model that can be used to determine the sensitivityof the experiment with minimal computational resources. The model is able to describe theelectromagnetic response of a closed dielectric haloscope, accounting for realistic geometricimperfections, as well as the noise introduced by the receiver system. It is applied to datataken with a MAgnetized Disk and Mirror Axion Experiment (MADMAX) prototype within the 1.6 TMorpurgo magnet at CERN. This work underpins the first axion dark matter search using adielectric haloscope and provides the foundation for future dark matter searches with MADMAX.

Ivanov, A. [Munich, Max Planck Inst. Quantenopt.]↗

Implementation of a multi-threaded framework for large-scale scientific applications

The CMS experiment has recently completed the development of a multi-threaded capable application framework. In this paper, we will discuss the design, implementation and application of this framework to production applications in CMS. For the 2015 LHC run, this functionality is particularly critical for both our online and offline production applications, which depend on faster turn-around times and a reduced memory footprint relative to before. These applications are complex codes, each including a large number of physics-driven algorithms. While the framework is capable of running a mix of thread-safe and 'legacy' modules, algorithms running in our production applications need to be thread-safe for optimal use of this multi-threaded framework at a large scale. Towards this end, we discuss the types of changes, which were necessary for our algorithms to achieve good performance of our multithreaded applications in a full-scale application. Lastly performance numbers for what has been achieved for the 2015 run are presented.

97 MATHEMATICS AND COMPUTING↗

Modelling the impact of quasar redshift errors on the full-shape analysis of correlations in the Lyman-α forest.

In preparation for the first cosmological measurements from the full shape of the Lyman-α (Lyα) forest from DESI, we must carefully model all relevant systematics that might bias our analysis. It was shown in Youles et al. (2022) that random quasar redshift errors produce a smoothing effect on the mean quasar continuum in the Lyα forest region. This, in turn, gives rise to spurious features in the Lyα autocorrelation and its cross-correlation with quasars. Using synthetic data sets based on the DESI survey, we confirm that the impact on BAO measurements is small, but that a bias is introduced to parameters which depend on the full shape of our correlations. We combine a model of this contamination in the cross-correlation (Youles et al. 2022) with a new model we introduce here for the auto-correlation. These are parametrised by 3 parameters, which, when included in a joint fit to both correlation functions, successfully eliminate any impact of redshift errors on our full-shape constraints. We also present a strategy for removing this contamination from real data, by removing ∼0.3% of correlating pairs.

cosmology↗

FitCache: A Transparent Drop-In Framework for Multi-Tier Caching to Accelerate Distributed Deep Learning Workloads

Training in Deep learning (DL) remains highly compute- and data-intensive, with I/O becoming a critical bottleneck as models and datasets scale. Recent studies report that data loading can dominate training time, especially on large-scale HPC systems with shared parallel file systems (PFS). Existing caching approaches either rely on single-tier designs or require intrusive modifications to training pipelines, limiting their portability and effectiveness. In this work, we present FitCache, a transparent drop-in framework for multi-tier caching to accelerate distributed DL training by coordinating fast local memory (e.g., DRAM, Persistent Memory (PMem)) and NVMe as hierarchical caches atop PFS. Our design adapts to hardware diversity, i.e., if NVMe is missing, memory transparently acts as a caching tier, ensuring stable performance. FitCache transparently intercepts I/O requests and issues concurrent fetches across all tiers, returning data from the fastest responder without centralized metadata or static redirection paths. FitCache adapts to dynamic workloads and heterogeneous clusters while maintaining POSIX compatibility. Experiments on Frontier (2048 GPUs) and smaller research clusters show that FitCache reduces training time by up to 40% and per-batch I/O latency by up to 71.6% compared to Lustre Orion PFS, offering a drop-in solution for scalable DL training.

Hu, Guangxing [ORNL] (ORCID:0009000283203614)↗

RADAI: A Large-Scale Realistic Dataset for Radiation Detection Algorithm Development

Open, realistic datasets are essential for developing and benchmarking radiation detection algorithms, yet they remain scarce. The Radiological Anomaly Detection and Identification (RADAI) project was develop to create datasets that meet the training and testing needs for sophisticated radiation detection algorithms. The RADAI dataset is a large-scale synthetic resource that integrates high-fidelity Monte Carlo simulations with realistic urban scenarios to capture both background variability and source signatures. RADAI models construction-material NORM, people and vehicles, urban clutter, and dynamic environmental effects such as cosmic-ray and rain-induced transients, and they provide list-mode detector data with motion and response modeling suitable for algorithm training and evaluation. The RADAI project resulted in three publicly-released complementary datasets together with an online scoring portal for standardized performance assessment and an open software toolkit that supports data access, augmentation, model development, and evaluation. These resources enable reproducible comparisons across methods and promote rigorous studies at the scale required by contemporary machine learning. By grounding algorithm development in realistic, well-documented conditions, RADAI supports progress toward more robust detection, identification, and localization in complex urban environments.

Ghawaly, James M. [Division of Computer Science an↗

Biomanufacturing and Scale-Up: Pathways to Biochemicals, Biofuels, and Biomaterials

Advancing the bioeconomy requires the development of large-scale microbial bioprocesses capable of converting waste carbon streams into biofuels, biochemicals, and biomaterials at industrially relevant scales. While biomanufacturing has been successfully demonstrated at the laboratory scale for a wide range of chemicals, only a few have reached industrial-scale production. This is partly due to the inherent complexity of microbial systems, which rely on living cells with intricate metabolic pathways that are highly sensitive to environmental changes, making large-scale production difficult to optimize and predict. As a result, scaling-up bioprocesses remains a high-stakes challenge that requires deeper exploration. This involves integrating feedstock and microbial selection, upstream and downstream processes, and computational modelling, among other research efforts. Bulk and specialty chemicals derived from biological processes also face competition from fossil-based production routes, which have been refined through decades of technological advancements. While biologically derived molecules may offer more environmentally friendly production pathways than traditional chemical manufacturing, their widespread adoption depends on achieving cost parity-or superiority-relative to fossil-based methods. This emphasizes the importance of holistic research, including techno-economic analyses and life cycle assessments, to ensure both economic viability and environmental sustainability. This editorial and special issue explores state-of-the-art strategies for converting waste carbon sources into valuable products. It discusses how enzymes, single microbes (e.g., extremophiles), and microbiomes (e.g., through division of labor) can be integrated with upstream and downstream process innovations-such as consolidated bioprocessing and in situ product recovery-to improve the efficiency and scalability of biomanufacturing. The editorial further highlights the role of computational modelling in understanding, predicting, and controlling bioprocess performance across scales, and concludes by emphasizing the importance of techno-economic modelling to identify technologies that can move to market.

09 BIOMASS FUELS↗

Algebraic Multigrid with Filtering: An Efficient Preconditioner for Interior Point Methods in Large-Scale Contact Mechanics Optimization

Large-scale contact mechanics simulations are crucial in many engineering fields such as structural design and manufacturing. In the frictionless case, contact can be modeled by minimizing an energy functional; however, these problems are often nonlinear, nonconvex, and increasingly difficult to solve as mesh resolution increases. In this work, we employ a Newton-based interior-point (IP) filter line-search method, an effective approach for large-scale constrained optimization. While this method converges rapidly, each iteration requires solving a large saddle-point linear system that becomes ill-conditioned as the optimization process converges, largely due to IP treatment of the contact constraints. Such ill-conditioning can hinder solver scalability and increase iteration counts with mesh refinement. Here, to address this, we introduce a novel preconditioner, algebraic multigrid with filtering (AMGF), tailored to the Schur complement of the saddle-point system. Building on the classical AMG solver, commonly used for elasticity, we augment it with a specialized subspace correction that filters near null space components introduced by contact interface constraints. Through theoretical analysis and numerical experiments on a range of linear and nonlinear contact problems, we demonstrate that the proposed solver achieves mesh independent convergence and maintains robustness against the ill-conditioning that notoriously plagues IP methods. These results indicate that AMGF makes contact mechanics simulations more tractable and broadens the applicability of Newton-based IP methods in challenging engineering scenarios. More broadly, AMGF is well suited for problems, optimization or otherwise, where solver performance is limited by a low-dimensional subspace, such as those arising from localized constraints, interface conditions, or model heterogeneities. This makes the method widely applicable beyond contact mechanics and constrained optimization.

Mathematics and Computing↗

GP Cosmology Surrogate v1.0

GP Cosmology Surrogate is a Python library for building and training a generalized multi-output Gaussian process (GP) framework of @takhtaganov2021cosmic. In this approach, the surrogate is constructed sequentially, guided by a Bayesian optimization acquisition function that targets reduction of emulation error in the regions most consistent with the observational data. This adaptive design concentrates computational resources where they have the greatest impact on inference accuracy. The library supports efficient training for separable GP kernels, which allows the use of Kronecker algebra to handle high-dimensional input spaces and large numbers of correlated outputs. This makes it well suited for applications such as modeling cosmological power spectra, large-scale physical simulations, and multi-output hyperparameter tuning. By combining scalable multi-output GP modeling with data-driven adaptive sampling, GPsurrogate enables parameter inference and optimization with substantially fewer simulations than conventional space-filling designs.

Lukic, Zarija [Lawrence Berkeley National Laborato↗

Enabling Exascale Hardware and Software Design through Scalable System Virtualization (Final Report)

In this grant, we enhanced the Palacios virtual machine monitor to increase its scalability and suitability for addressing exascale system software design issues. This included a wide range of research on core Palacios features, large-scale system emulation, fault injection, perfomrance monitoring, and VMM extensibility. This research resulted in large number of high-impact publications in well-known venues, the support of a number of students, and the graduation of two Ph.D. students and one M.S. student. In addition, our enhanced version of the Palacios virtual machine monitor has been adopted as a core element of the Hobbes operating system under active DOE-funded research and development.

97 MATHEMATICS AND COMPUTING↗

Proposed New STELLA-2 Tests

In 2025, KAERI and Argonne initiated a collaboration based on software validation using data from tests performed at KAERI’s STELLA-2 facility. STELLA-2 is a large-scale sodium thermal-hydraulic test facility that was originally developed to support KAERI’s development of the PGSFR design concept. KAERI has performed a large number of tests at STELLA-2, providing valuable data to support sodium fast reactor code validation. KAERI will be conducting new tests in 2026, targeting system conditions that were not achieved during previous test campaigns. KAERI has requested from Argonne proposed tests that could be performed during the upcoming test campaign. This report documents Argonne’s proposed tests in fulfillment of KAERI’s request.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

Confinement Exploiting Arrays of Cross-Flow Turbines (ConExT) (Final Scientific/Technical Report)

Final technical report for the ARPA-E SHARKS project: Confinement Exploiting Arrays of Cross-Flow Turbines (ConExT). This involved collaboration by the University of Washington, University of Wisconsin, National Renewable Energy Laboratory, and Oberon Insights. Cost-effective, large-scale utilization of tidal and river current resources requires turbine arrays. Most array concepts involve multiple, staggered rows of turbines analogous to wind farms. However, in water, turbines that have an appreciable projected area (i.e., the projected area over a full revolution) relative to a channel’s cross-sectional area can theoretically extract far more energy than when operating in isolation. This suggests substantial benefits to a confinement-exploiting approach to array layout, in which turbines are more densely clustered in a single row spanning a channel. The objective of this project was to move this concept from theory to practice, as well as establishing environmental and economic trade-offs. This objective was accomplished through scale-model experimentation, high-fidelity computational fluid dynamic simulation, and integrated techno-economic modeling.

16 TIDAL AND WAVE POWER↗

A Large-Scale Hardware Experiment Demonstration of Operating High Inverter-Based Resource Power Systems With Grid-Forming Inverters

This paper presents experimental hardware results from a microgrid system as the penetration level of grid-forming (GFM) inverters increases. The experiment aims to showcase the advantages of GFM inverters in enhancing system stability and to investigate the operational challenges in systems relying entirely on inverter-based resources (IBRs). Five test scenarios are devised to progressively increase GFM inverter penetration levels: 0% (S1), 19% (S2), 37% (S3), 68% (S4), and 100% (S5). To ensure consistency, identical loading conditions and dynamic events are applied across all scenarios. The conducted tests include load step changes, output variation of grid-following (GFL) inverters during islanded mode, transition operations (such as synchronization to the grid and islanding), and rateof- change-of-frequency (ROCOF) and voltage jump tests during grid-connected mode. Key findings from the tests are summarized as follows: (1) Scenario S1 proved most challenging for load steps, and GFL variations because of insufficient GFM capacity, leaving the diesel generator unable to handle transient ridethrough; (2) Scenario S5 was the most difficult for transition operations, because the lack of a diesel generator to maintain stiff bus voltage resulted in unexpected reactive power flows during synchronization, causing the point of common coupling (PCC) breaker to trip; (3) Scenarios S4 and S5 were particularly challenging for ROCOF tests because of minimal system inertia, leading to large transients that triggered breaker trips; and (4) Scenario S4 exhibited the strongest voltage recovery capability because of the presence of the diesel generator and a higher number of GFM inverters, both of which possess the highest capacity for reactive power injection.

24 POWER TRANSMISSION AND DISTRIBUTION↗

ODIN: Probing the LAE Ly α Luminosity Function across Cosmic Time and Different Environments

The ubiquity and relative ease of discovery make 2 ≲ z ≲ 5 Lyα emitting galaxies (LAEs) ideal tracers for large-scale structure of the distant Universe. In addition, because Lyα is a resonance line, but frequently observed at large equivalent width, it is potentially a probe of galaxy evolution. The LAE Lyα luminosity function (LF) is an essential measurement for making progress on both of these topics. Although several studies have computed the LAE LF, very few have delved into how the function varies with environment. The large area and depth of the One-hundred-deg 2 DECam Imaging in Narrowbands (ODIN) survey makes such measurements possible at the cosmic noon redshifts of z ∼ 2.4, 3.1, and 4.5. In this initial work, we present algorithms needed to rigorously compute the LAE LF, and test them on the ∼16,000 ODIN LAEs found in the extended COSMOS field. Using these limited samples, we find weak evidence that protocluster environments suppress the numbers of faint LAEs compared to the field. We also find that the LF decreases in number density and evolves towards a steeper faint-end slope over cosmic time from z ∼ 4.5 to z ∼ 2.4.

Lyman-alpha galaxies↗

On the Properties of Cosmological Ionization Fronts

We investigate the properties of cosmological ionization fronts during the epoch of reionization using the Cosmic Reionization on Computers simulations. By analyzing reionization timing maps, we characterize ionization front velocities and curvatures and their dependence on the density structure of the intergalactic medium (IGM). The velocity distribution of ionization fronts in the simulations indicates that while the barrier-crossing analytical model captures the overall shape in high-velocity regions, it fails to reproduce the low-velocity tail, highlighting the non-Gaussian nature of the IGM’s density field. Ionization front velocities are inversely correlated with local density, propagating faster in underdense regions and more slowly in overdense environments. Faster ionization fronts also lead to higher postionization temperatures, reaching a plateau at ∼2 × 10$^{4}$ K for velocities exceeding 3000 km s$^{−1}$. Examining curvature statistics further establishes a connection between ionization front structure and the normalized density contrast ν, with trends in overdense regions aligning well with barrier-crossing model predictions, while deviations appear in underdense environments due to model limitations. These results provide a detailed characterization of ionization front dynamics and their interaction with the underlying density field, bridging small-scale reionization physics with large-scale observables such as the 21 cm signal and the IGM’s thermal history.

79 ASTRONOMY AND ASTROPHYSICS↗

High-Resolution ESM Projections for Energy Applications Over the CONUS

Assessing energy resources under future scenarios requires high-resolution meteorological information that is physically consistent and suitable for regional-scale analysis. While Earth system model (ESM) projections provide valuable large-scale information, their coarse resolution and systematic biases limit direct applicability for energy system modeling and planning. In this study, we develop a high-resolution dynamical downscaling framework based on the Weather Research and Forecasting (WRF) model to translate global-scale ESM data into energy-relevant regional projections over the contiguous United States (CONUS). The framework identifies an optimized WRF configuration through numerical experiments and evaluates raw and bias-corrected ESM initial and boundary conditions, with soil moisture (SM) and soil temperature (ST) bias correction implemented as an integral part of the bias-corrected ESM forcing to improve land-atmosphere coupling prior to WRF dynamical downscaling. Using an optimized WRF configuration at 4-km resolution, we show that raw ESM forcing introduces systematic dry and cold soil biases that propagate into pronounced warm biases in near-surface air temperature and positive biases in solar irradiance, particularly during summer. Applying bias-corrected atmospheric forcing together with bias-corrected SM and ST substantially reduces these downstream biases and improves the surface energy balance and near-surface atmospheric fields. These results demonstrate that bias-aware treatment of initial conditions is critical for producing high-resolution downscaled projections suitable for energy system modeling and planning applications.

24 POWER TRANSMISSION AND DISTRIBUTION↗

There and Back Again: Reimagining Cryogenic Cooling for Scalable Arrays of Dilution Refrigerators for future Quantum Datacenters

While pulse tube cryocoolers enabled the rapid expansion of dilution refrigerator technology over the past two decades, the transition to large-scale quantum systems is now driving a reassessment of the DR’s higher-temperature-stage cooling strategies and how these systems can be effectively scaled in a modular way. Quasi-wet architectures based on centralized cryoplants and forced-flow helium distribution offer compelling advantages in energy efficiency, operational cost, and scalability. With appropriate redundancy, standardized interfaces, and optimized distribution system designs, these architectures will provide a practical and robust path forward for the next generation of quantum computing infrastructure.

Hansen, B. [Fermilab]↗