Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “parallel data structures”

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 451 records · Page 25
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Numerical and experimental analysis of mechanically induced failure in electric vehicle battery modules

Mitigating thermal runaway and cell-to-cell propagation is essential for improving the safety of electric and hybrid vehicles. Enhancing digital twin capabilities to predict battery mechanical abuse is particularly critical for automotive and aerospace applications, where crashworthiness is a key concern. Understanding failure conditions and propagation in battery modules during mechanical abuse is complex due to interactions between structural deformation, heat transfer, electrochemical processes, exothermic reactions and mechanical fracture. While prior studies have focused on modeling cell-level behavior, extending these models to module or pack level is necessary for a system level understating of electric vehicle safety. This study develops coupled large deformation finite element models that simultaneously solve for electrochemistry, material failure, internal short circuit and thermal runaway propagation. The models account for mechanical and thermal interactions between lithium-ion cells and other battery components while the contact interfaces are evolving with time. Model-predicted voltage, temperature and force responses are compared with experimental data for validation. The results demonstrate that the approach captures key failure mechanisms, including thermal propagation through heat transfer, electrical propagation from short circuits in parallel-connected cells, and mechanical propagation via penetration and crack formation. These findings show that computational models are valuable tools for understanding battery module failure and providing insight that can reduce the need for extensive experimental testing.

25 ENERGY STORAGE↗
DOE OSTI2026

Explainable machine learning for incipient anomaly detection in compact molten salt heat exchanger with overlapping feature distributions

High-temperature molten salt-cooled reactors (MSCRs) are a promising next-generation nuclear technology option, offering efficient power conversion and inherent safety features. However, the reliability of these systems depends on the robust operation of heat exchangers (HXs), which are susceptible to failure due to temperature gradients and channel plugging caused by fluid freezing. Conventional monitoring methods, relying on inlet and outlet measurements, lack the spatial resolution needed to detect early-stage faults. We propose a novel design of a compact salt-to-salt matrix-type HX design consisting of interleaved arrays of parallel tubes, with integrated synthetic fiber optic distributed temperature sensing (DTS) to enable localized detection of incipient faults. To evaluate performance of this design, we generate high-fidelity synthetic data using heat transfer computational modeling to simulate channel plugging, and introduce sensor noise for realistic modeling of measurements. The dataset comprises of 97% normal operation and 3% anomaly cases, with each anomaly class representing 1% of the data. These early anomalies result in overlapping temperature profiles between normal and faulty channels, producing a non-separable dataset that challenges traditional classification techniques. We benchmark eight supervised machine learning (ML) models and demonstrate that XGBoost achieves the highest performance. To improve transparency, we develop an explainability framework combining Shapley values and partially ordered sets (POSETs) to quantify and structurally analyze feature importance. This approach identifies both dominant predictors and ambiguous feature relationships, enhancing trust and interpretability. Our results highlight the potential of combining DTS and explainable ML with intelligent feature selection to improve predictive maintenance and ensure operational resilience in advanced nuclear systems.

Prantikos, Konstantinos [Argonne National Laborato↗
DOE OSTI2026

PyJMAK: An Open-Source Python Toolkit for Modeling Solid-State Metallurgical Phase Transformations

Accurate prediction of metallurgical phase transformations is an essential basis for autonomous optimization and rapid part qualification. Several methods can be used to estimate the evolution of phase fractions such as JMAK kinetics-based models, phase-field models, thermodynamic models, and data-driven machine learning models. Thermodynamic and phase-field-based methodologies solve multiphysics equations requiring numerous calibration parameters and significant computational resources. As a result, the computation domain is limited to a point or on order of micron-meters. The data-driven models rely on large datasets from experiments and simulations. While the JMAK model only provides information about phase fraction evolution, it can predict this evolution in near real-time using thermal history and thermodynamic data without restriction on the domain. JMAK models have been popularly used by researchers to model phase transformations occuring during additive manufacturing or over arbitrary temperature profiles. Commercial proprietary software such as Abaqus and Ansys or closed-source in-house implementations offer the ability to model JMAK based kinetics to predict phase transformation. However, these software packages are not open-source or freely available for use and development in conjunction with manufacturing machines, sensors, and machine learning algorithms. In addition, the use of the model is restricted by a license token. In contrast, given temperature profiles at multiple points in the domain, this Python-based PyJMAK model can compute phase evolution in parallel due to its stand-alone modular, voxel-based structure, and it can be executed on high-performance computing resources without any license restrictions.

Prabhune, Bhagya [Oak Ridge National Laboratory (O↗
DOE OSTI2026

Thermal Performance of Spandrel Assemblies in Glazed Wall Systems: Laboratory Test Design – Challenges and Test Results

Accurate thermal performance calculation procedures for opaque spandrel areas in curtain wall and window wall systems are essential for rating systems when comparing spandrel systems. However, there is a lack of consensus in thermal modeling needed for accurately characterizing heat transfer through spandrel assemblies due to the complex arrangement of materials and structural components. Several studies indicate that conventional 2D thermal simulations may overestimate R-values by 30% compared to physical testing and 3D simulations. Detailed simulations and well-curated laboratory test data are necessary to build confidence in simulation models, which will later be used to develop correlations to improve widely used conventional 2D thermal simulations. This study aims to experimentally test heat transfer through various spandrel assemblies to validate 3D simulation models. Also, the challenges of conducting a thorough testing design along with the solutions would be documented. The team developed a design for testing spandrel assemblies, making appropriate modifications to the existing heat, air, and moisture (HAM) chamber to accommodate the testing needs. Two moveable baffles were designed and fabricated to guide airflow direction parallel to the test article surface. The data acquisition capabilities in the chamber were upgraded to add more than two hundred sensors to the climate and indoor side of the chamber. The goal is to provide a quality dataset for validating complex 3D modeling simulations, which will be used to develop improved thermal simulation techniques that more accurately represent the thermal behavior of spandrel assemblies and their integration within the building envelope. This paper will summarize the results for the boundary conditions of the testing and the temperature variation across different locations of the spandrel assemblies.

Kunwar, Niraj [ORNL] (ORCID:0000000263457652)↗
DOE OSTI2024

Grain structure and texture selection regimes in metal powder bed fusion

Additive manufacturing (AM) offers opportunities to produce complex part geometries not possible with conventional processing and in some cases even improve part performance. However, adoption has been slowed by difficulties assessing microstructure variability and there is no straightforward approach to relate processing to grain structure characteristics. In this study, datasets from AdditiveFOAM heat transport simulations of laser powder bed fusion (LPBF) are used to drive ExaCA simulations of grain structure. The GPU utilization of ExaCA and an algorithmic update for modeling melt pool overlap region solidification enabled rapid and parallel simulation across a wider range of process conditions than previously explored with cellular automata-based solidification models. A texture selection angle $θ_s$ is defined based on melt pool overlap geometry, and the range of $θ_s$ over which a commonly observed texture transition occurs in characterized AM builds was well-reproduced by ExaCA simulations over a wide range of melt pool shape, hatch spacing, and layer height. ExaCA simulations with 90 degree rotation of the scan direction on every other layer reproduced a number of trends from the AM literature including grain refinement, the dominance of layers with larger melt pools on the final grain structure, and the weakening or strengthening of texture depending on odd and even layer melt pool overlap geometry. EBSD data from a benchmark AM part is used to validate the simulated mechanism of a layer rotation-induced texture strengthening effect. Importantly, these results expand the understanding of the mechanisms for texture selection in alloys with cubic crystal symmetry and offer an approach to easily evaluate processing conditions. With this new understanding, these modeling tools will enable anticipation of previously unexpected variations in grain structure and target specific microstructures and properties.

36 MATERIALS SCIENCE↗
DOE OSTI2020

A Single Small-Scale Plutonium Redox Reaction System Yields Three Crystallographically-Characterizable Organoplutonium Complexes

Here, an approach to obtaining substantial amounts of data from a hazardous starting material that can only be obtained and handled in small quantities is demonstrated by the investigation of a single small-scale reaction of cyclooctatetraene, C 8 H 8 , with a solution obtained from the reduction of Cp' 3 Pu (Cp' = C 5 H 4 SiMe 3 ) with potassium graphite. This one reaction coupled with oxidation of a product has provided single-crystal X-ray structural data on three organoplutonium compounds as well as information on redox chemistry thereby demonstrating an efficient route to new reactivity and structural information on this highly radioactive element. The crystal structures were obtained from the reduction of C 8 H 8 by a putative Pu(II) complex, (Cp' 3 Pu II ) 1– , generated in situ, to form the Pu(III) cyclooctatetraenide complex, [K(crypt)][(C 8 H 8 ) 2 Pu III ], 1-Pu , and the tetra(cyclopentadienyl) Pu(III) complex, [K(crypt)][Cp' 4 Pu III ], 2-Pu . Oxidation of the sample of 1-Pu with Ag(I) afforded a third organoplutonium complex that has been structurally characterized for the first time, (C 8 H 8 ) 2 Pu IV , 3-Pu . Complexes 1-Pu and 3-Pu contain Pu sandwiched between parallel (C 8 H 8 ) 2– rings. The (Cp' 4 Pu III ) – anion in 2-Pu features three η 5 -Cp' rings and one η 1 -Cp' ring, which is a rare example of a formal Pu–C η 1 -bond. In addition, this study addresses the challenge of small-scale synthesis imparted by radiological and material availability of transuranium isotopes, in particular that of pure metal samples. A route to an anhydrous Pu(III) starting material from the more readily available Pu IV O 2 was developed to facilitate reproducible syntheses and allow complete spectroscopic analysis of 1-Pu and 2-Pu . Pu IV O 2 was converted to Pu III Br 3 (DME) 2 (DME = CH 3 OCH 2 CH 2 OCH 3 ) and subsequently Pu III Br 3 (THF) x , which was used to independently synthesize 1-Pu , 2-Pu , and 3-Pu .

38 RADIATION CHEMISTRY, RADIOCHEMISTRY, AND NUCLEA↗
DOE OSTI2025

Scale-up Unlearnable Examples Learning with High-performance Computing

Recent advancements in AI models, like ChatGPT, are structured to retain user interactions, which could inadvertently include sensitive healthcare data. In the healthcare field, particularly when radiologists use AI-driven diagnostic tools hosted on online platforms, there is a risk that medical imaging data may be repurposed for future AI training without explicit consent, spotlighting critical privacy and intellectual property concerns around healthcare data usage. Addressing these privacy challenges, a novel approach known as Unlearnable Examples (UEs) has been introduced, aiming to make data unlearnable to deep learning models. A prominent method within this area, called Unlearnable Clustering (UC), has shown improved UE performance with larger batch sizes but was previously limited by computational resources (e.g., a single workstation). To push the boundaries of UE performance with theoretically unlimited resources, we scaled up UC learning across various datasets using Distributed Data Parallel (DDP) training on the Summit supercomputer. Our goal was to examine UE efficacy at high-performance computing (HPC) levels to prevent unauthorized learning and enhance data security, particularly exploring the impact of batch size on UE’s unlearnability. Utilizing the robust computational capabilities of the Summit, extensive experiments were conducted on diverse datasets such as Pets, MedMNist, Flowers, and Flowers102. Our findings reveal that both overly large and overly small batch sizes can lead to performance instability and affect accuracy. However, the relationship between batch size and unlearnability varied across datasets, highlighting the necessity for tailored batch size strategies to achieve optimal data protection. The use of Summit’s high-performance GPUs, along with the efficiency of the DDP framework, facilitated rapid updates of model parameters and consistent training across nodes. Our results underscore the critical role of selecting appropriate batch sizes based on the specific characteristics of each dataset to prevent learning and ensure data security in deep learning applications. The source code is publicly available at https: // github. com/ hrlblab/ UE_ HPC .

Zhu, Yanfan [Vanderbilt University, Nashville, TN,↗
DOE OSTI2022

Development of a high fidelity model of the CROCUS experimental reactor

Measurements of scalar flux distributions with fine spatial and energy resolutions are needed to remedy one of the validation shortcomings of the novel neutronics full core solvers, such as MPACT and nTRACER. While a very detailed resolution of the flux can be calculated with such codes, only a limited experimental data set is available to check their accuracy. Such type of measurements are on-going at the zero power reactor CROCUS, operated at the Laboratory for Reactor Physics and System Behaviours of the EPFL, thanks to the development of advanced miniature neutron detection systems. This kind of experimental data would provide the community with a suitable benchmark for the validation of high fidelity neutronics solvers. In parallel, a multi-physics solver for steady-state and transient analysis of nuclear reactors, named GeN-Foam, has been developed. Based on the finite-volume OpenFOAM library, GeN-Foam provides us with enough flexibility to analyze non-conventional reactor geometries such as that of CROCUS. While CROCUS heterogeneities cannot be modeled by MPACT and nTRACER for the moment, GeN-Foam offers a unique opportunity to build a high fidelity model which mimics these codes' method to reach sub-pin simulation resolution. This document aims at describing the work achieved to get from the existing GeN-Foam model of the CROCUS reactor based on a structured mesh and using the neutron diffusion, the first high-fidelity model using discrete ordinates method as an approximation to neutron transport and an unstructured mesh for inter lattice water gap description and sub-pin heterogeneous modeling. (authors)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗
DOE OSTI2021

Visual HPC Workflows for the Analysis of System Dynamics Models

Visual analytics supported by high performance computing (HPC) accelerates and enhances the discovery, exploration, and analysis of causal patterns in complex system dynamics (SD) models. We present a suite of visualization-assisted ensemble-based techniques for hypothesis generation and testing, and for sensitivity analysis. By employing HPC to provide parallel, on-demand simulation of SD models, one can “steer” an ensemble of simulated scenarios in real time as one first formulates and then informally tests those hypotheses: this provides rapid feedback for analysts to refine their understanding of the causal relationships emergent from a model. Such understandings can be followed and augmented by rigorous application of statistical methods, namely global variance-based sensitivity analysis, Monte-Carlo filtering, adaptive regional sensitivity analysis, and self-organized maps: here timely computation relies on HPC, while effective presentation emphasizes high-dimensional multivariate data visualization. Immersive visualization in virtual 3D environments provides an excellent adjunct to the traditional 2D graphics typically used for SD models, as it generates an embodied understanding of model behavior and facilitates an active, collaborative critique of model structure and output. Finally, we summarize prospects for HPC-enabled visual analytics applied to SD modeling.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2017

Phenomenological Model of Hydrophobic and Hydrophilic Interactions

Hydration forces acting between macroscopic bodies at distances L ≤ 3 nm in pure water are calculated based on the phenomenological model of polar liquids. It is shown that depending on the properties of the bodies, the interacting surfaces polarize the liquid differently, and wetting properties of the surfaces are completely characterized by two parameters. If the surfaces are hydrophilic, liquid molecules are polarized at right angles to the surfaces, and the interaction is the short-range repulsion (the forces of interaction decrease exponentially over the characteristic length λ ≈ 0.2 nm). The interaction between the hydrophobic surfaces is more diversified and has been studied less. For L ≤ 3 nm, the interaction exhibits universal properties, while for L ≤ 3 nm, it considerably depends on the properties of the surfaces and on the distances between them, as well as on the composition of the polar liquid. In full agreement with the available experimental results we find that if the interfaces are mostly hydrophobic, then the interaction is attractive and long-range (the interaction forces diminish exponentially with decay length 1.2 nm). In this case, the resultant polarization of water molecules is parallel to the surface. It is shown that hydration forces are determined by nonlinear effects of polarization of the liquid in the bulk or by analogous nonlinearity of the interaction of water with a submerged body. This means that the forces of interaction cannot be calculated correctly in the linear response approximation. The forces acting between hydrophobic or hydrophilic surfaces are of the entropy type or electrostatic, respectively. It is shown that hydrophobic and hydrophilic surfaces for L ≤ 3 nm repel each other. The calculated intensity of their interaction is in agreement with experimental data. We predict the existence of an intermediate regime in which a body cannot order liquid molecules, which results in a much weaker attraction that decreases in inverse proportion to the squared distance between the surfaces of bodies. The difference between the microscopic structures of liquids confined in nanovolumes from liquids in large volumes is considered. The proposed model is applicable for a quantitative description of the properties of water at temperatures T satisfying the condition 0 < (T–T{sub c})/T{sub c} ≪ 1, where T{sub c} ≈ 230 K is the temperature of the ferroelectric phase transition observed in supercooled water. Under standard conditions, the model can be used for obtaining qualitative estimates.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2019

Modern Integrated Decommissioning at Chalk River Laboratories - 19401

Canadian Nuclear Laboratories (CNL), formerly Atomic Energy of Canada Limited (AECL) is Canada's premier nuclear science and technology (S and T) organization in Canada. Over the next decade, CNL will be transforming its many different sites through the revitalization of essential site infrastructure, the decommissioning of aging infrastructure and a significant investment in new, world-class science facilities. In particular, the main site for CNL resides at the Chalk River Laboratories (CRL), where the majority of parallel activities of revitalization and decommissioning will take place. In 2016, CRL started an ambitious ten year plan to decommission more than 120 buildings. In order to ensure this project remains on track and the ALARA (as low as reasonably achievable) principle is maintained, a strategy is required. This strategy is to focus on integrated activities, which meet/exceed international best practices, to ensure the work is carried out as efficiently as possible. Given the complexity of CRL operations and the legacy issues accumulated over 60 years of operation, CNL needed to adopt a modern and integrated approach to handle competing project priorities. As a starting point, achieving complete radiological characterization of many building structures is required. CNL engaged various vendors to modernize radiological surveys from paper records to electronic documentation, deploying modern survey tools such as data-logging instruments, gamma cameras, robotic surveying tools, and introducing tablets to improve the flow of information while workers are in the field. Once fully implemented, these improvements will ensure that quality information is readily available at all stages of decommissioning, which greatly reduces project schedules and results in important company savings. These techniques will ensure that, during planning, historic radiological survey data is easily searchable from the electronic documentation system. Any additional data required can be quickly and safety collected using robotic surveying tools in areas of higher radiation to reduce possible exposure to individuals. As decommissioning progresses, individuals in the field can use tablets, gamma cameras and data-logging instruments to accurately retrieve, collect and share data. As a result of these efforts, CNL is progressing towards becoming a leader in the field of decommissioning within Canada, applying modern techniques and best practices to reduce decommissioning project risk. Access to high quality information results in improved decommissioning plans, which will reduce the amount of rework and scheduling delays required and improve worker safety as the project is executed. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2023

Thermal expansion of 4H and 6H SiC from 5 K to 340 K

The first thermal expansion measurements of the 4H and 6H polytypes of SiC below room temperature are reported. The measurements were carried out on single-crystal specimens using high-resolution capacitive-based dilatometry. For both polytypes, the thermal expansion coefficient is below 2.4 × 10 -6 1/K near room temperature. No phase transitions are observed over the 5 K to 340 K temperature range of the measurements. The thermal expansion coefficient α of 4H SiC is slightly anisotropic for measurements parallel and perpendicular to the crystallographic c axis with α ∥ about 2.2 × 10 -7 1/K larger than α ⟂ near room temperature. For 6H SiC no discernible anisotropy is observed. The differences in anisotropy can be understood by considering the ratio of hexagonal to cubic bonds of each polytype. Narrow regions with negative thermal expansion that are within the limits of our resolution (~ 1 x 10 -8 ) are observed in the vicinity of 30 K for both specimens. In conclusion, tabulated data, polynomial fits, fit parameters, and comparison to data based on lattice-parameter measurements are provided.

36 MATERIALS SCIENCE↗
DOE OSTI2019

Lynx x-ray microcalorimeter

Lynx is an x-ray telescope, one of four large satellite mission concepts currently being studied by NASA to be a flagship mission. One of Lynx’s three instruments is an imaging spectrometer called the Lynx x-ray microcalorimeter (LXM), an x-ray microcalorimeter behind an x-ray optic with an angular resolution of 0.5 arc sec and ~2 m 2 of area at 1 keV. The LXM will provide unparalleled diagnostics of distant extended structures and, in particular, will allow the detailed study of the role of cosmic feedback in the evolution of the Universe. We discuss the baseline design of LXM and some parallel approaches for some of the key technologies. The baseline sensor technology uses transition-edge sensors, but we also consider an alternative approach using metallic magnetic calorimeters. We discuss the requirements for the instrument, the pixel layout, and the baseline readout design, which uses microwave superconducting quantum interference devices and high-electron mobility transistor amplifiers and the cryogenic cooling requirements and strategy for meeting these requirements. For each of these technologies, we discuss the current technology readiness level and our strategy for advancing them to be ready for flight. We also describe the current system design, including the block diagram, and our estimate for the mass, power, and data rate of the instrument.

47 OTHER INSTRUMENTATION↗
DOE OSTI2017

Study of the second magnetization peak and the pinning behaviour in Ba(Fe 0.935 Co 0.065 ) 2 As 2 pnictide superconductor

Isothermal magnetic field dependence of magnetization and magnetic relaxation measurements were performed for the $H\parallel {\rm{c}}$ axis on a single crystal of Ba(Fe 0.935 Co 0.065 ) 2 As 2 pnictide superconductor having T c = 21.7 K. The second magnetization peak (SMP) for each isothermal M(H) was observed in a wide temperature range from T c to the lowest temperature of measurement (2 K). The magnetic field dependence of relaxation rate R(H), showed a peak (H spt) between H on (onset of SMP in M(H)) and H p (peak field of SMP in M(H)), which is likely to be related to a vortex-lattice structural phase transition, as suggested in the literature for a similar sample. In addition, the magnetic relaxation measured for magnetic fields near H spt showed some noise, which might be the signature of the structural phase transition of the vortex lattice. Analysis of the magnetic relaxation data using Maley's criterion and the collective pinning theory suggested that the SMP in the sample was due to the collective (elastic) to plastic creep crossover, which was also accompanied by a rhombic to square vortex lattice phase transition. The analysis of the pinning force density suggested a single dominating pinning mechanism in the sample, which did not showing the usual $\delta {l}$ and $\delta {T}_{{\rm{c}}}$ nature of pinning. Furthermore, the critical current density (J c ), estimated using the Bean critical state model, was found to be 5 × 10 5 A cm - 2 at 2 K in the zero magnetic field limit. Surprisingly, the maximum of the pinning force density was not responsible for the maximum value of the critical current density in the sample.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2022

Quantitative spatiotemporal mapping of thermal runaway propagation rates in lithium-ion cells using cross-correlated Gabor filtering

Abuse testing of lithium-ion batteries is widely performed in order to develop new safety standards and strategies. However, testing methodologies are not standardised across the research community, especially with failure mechanisms being inherently difficult to reproduce. High-speed X-ray radiography is proven to be a valuable tool to capture events occurring during cell failure, but the observations made remain largely qualitative. We have therefore developed a robust image processing toolbox that can quantify, for the first time, the rate of propagation of battery failure mechanisms revealed by high-speed X-ray radiography. Using Gabor filter, the toolbox selectively tracks the electrode structure at the onset of failure. This facilitated the estimation of the displacement of electrodes undergoing abuse via nail penetration, and also the tracking of objects, such as the nail, as it propagates through a cell. Further, by cross-correlating the Gabor signals, we have produced practical, illustrative spatiotemporal maps of the failure events. From these, we can quantify the propagation rates of electrode displacement prior to the onset of thermal runaway. The highest recorded acceleration (≈514 mm s -2 ) was when a nail penetrated a cell radially (perpendicular to the electrodes) as opposed to axially (parallel to the electrodes). The initiation of thermal runaway was also resolved in combination with electrode displacement, which occurred at a lower acceleration (≈108 mm s -2 ). Our assistive toolbox can also be used to study other types of failure mechanisms, extracting otherwise unattainable kinetic data. Ultimately, this tool can be used to not only validate existing theoretical mechanical models, but also standardise battery failure testing procedures.

25 ENERGY STORAGE↗
DOE OSTI2025

A rheological model for loose sands with insights from DEM

A rheological model for loose granular media is developed to capture both solid-like and fluid-like responses during shearing. The proposed model is built by following the mathematical structure of an extended Kelvin–Voigt model, where an elastic spring and plastic slider act in parallel to a viscous damper. This arrangement requires the partition of the total stress into rate-independent and rate-dependent stress components. To model the solid-like behavior, a simple frictional plasticity model is adopted without modifications, thus contributing to the rate-independent stress. Instead, the fluid-like or rate-dependent stress is further decomposed into deviatoric and volumetric parts, by proposing a new formulation based on a combination of the μ(I) relation, originally developed under pressure-controlled shear, with a pressure-shear rate relation derived under volume-controlled shear. The proposed formulation allows the model to capture both the increase in the friction coefficient and the enhanced dilation at high shear rates. High-fidelity simulation data, obtained from discrete element method and multiscale modelling, are used to evaluate the performance of the proposed constitutive model. The model provides accurate results under both drained and undrained simple shear paths across a wide range of shear rates. Furthermore, it successfully reproduces at much lower computational cost the flowslide mobility computed through multiscale simulations, which is primarily regulated by the shear rate dependence of the material properties during the dynamic runout stage.

Elasticity↗
DOE OSTI2024

Regional-scale fault-to-structure earthquake simulations with the EQSIM framework: Workflow maturation and computational performance on GPU-accelerated exascale platforms

Continuous advancements in scientific and engineering understanding of earthquake phenomena, combined with the associated development of representative physics-based models, is providing a foundation for high-performance, fault-to-structure earthquake simulations. However, regional-scale applications of high-performance models have been challenged by the computational requirements at the resolutions required for engineering risk assessments. The EarthQuake SIMulation (EQSIM) framework, a software application development under the US Department of Energy (DOE) Exascale Computing Project, is focused on overcoming the existing computational barriers and enabling routine regional-scale simulations at resolutions relevant to a breadth of engineered systems. This multidisciplinary software development—drawing upon expertise in geophysics, engineering, applied math and computer science—is preparing the advanced computational workflow necessary to fully exploit the DOE’s exaflop computer platforms coming online in the 2023 to 2024 timeframe. Achievement of the computational performance required for high-resolution regional models containing upward of hundreds of billions to trillions of model grid points requires numerical efficiency in every phase of a regional simulation. This includes run time start-up and regional model generation, effective distribution of the computational workload across thousands of computer nodes, efficient coupling of regional geophysics and local engineering models, and application-tailored highly efficient transfer, storage, and interrogation of very large volumes of simulation data. This article summarizes the most recent advancements and refinements incorporated in the workflow design for the EQSIM integrated fault-to-structure framework, which are based on extensive numerical testing across multiple graphics processing unit (GPU)-accelerated platforms, and demonstrates the computational performance achieved on the world’s first exaflop computer platform through representative regional-scale earthquake simulations for the San Francisco Bay Area in California, USA.

58 GEOSCIENCES↗
DOE OSTI2024

Optimizing High-Throughput Inference on Graph Neural Networks at Shared Computing Facilities with the NVIDIA Triton Inference Server

Abstract With machine learning applications now spanning a variety of computational tasks, multi-user shared computing facilities are devoting a rapidly increasing proportion of their resources to such algorithms. Graph neural networks (GNNs), for example, have provided astounding improvements in extracting complex signatures from data and are now widely used in a variety of applications, such as particle jet classification in high energy physics (HEP). However, GNNs also come with an enormous computational penalty that requires the use of GPUs to maintain reasonable throughput. At shared computing facilities, such as those used by physicists at Fermi National Accelerator Laboratory (Fermilab), methodical resource allocation and high throughput at the many-user scale are key to ensuring that resources are being used as efficiently as possible. These facilities, however, primarily provide CPU-only nodes, which proves detrimental to time-to-insight and computational throughput for workflows that include machine learning inference. In this work, we describe how a shared computing facility can use the NVIDIA Triton Inference Server to optimize its resource allocation and computing structure, recovering high throughput while scaling out to multiple users by massively parallelizing their machine learning inference. To demonstrate the effectiveness of this system in a realistic multi-user environment, we use the Fermilab Elastic Analysis Facility augmented with the Triton Inference Server to provide scalable and high-throughput access to a HEP-specific GNN and report on the outcome.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗