Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Heterogeneous 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 505 records · Page 28
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Size-Transferable Prediction of Excited State Properties for Molecular Assemblies with a Machine Learning Exciton Model

Computational modeling of the excited states of molecular aggregates faces significant computational challenges and size heterogeneity. Current machine learning (ML) models, typically trained on specific-sized aggregates, struggle with scalability. We found that the exciton model Hamiltonian of large aggregates can be decomposed into dimer pairs, allowing an ML model trained on dimers to reconstruct Hamiltonians for aggregates of any size. We also proposed a new method to address the phase-correction problem by introducing coupling terms’ approximations. Our model accurately predicted the excitation energies of the trimer and tetramer of perylene and tetracene and estimated S1 oscillator strengths of perylene aggregates. Leveraging our ML model, the optical gaps of nanosized perylene aggregates with up to 50 monomers are analyzed, qualitatively revealing the role of different couplings on their size dependency. Future work will explore transferability across different monomers to predict optical properties in heterogeneous assemblies.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2023

Highly active, ultra-low loading single-atom iron catalysts for catalytic transfer hydrogenation

Highly effective and selective noble metal-free catalysts attract significant attention. Here, a single-atom iron catalyst is fabricated by saturated adsorption of trace iron onto zeolitic imidazolate framework-8 (ZIF-8) followed by pyrolysis. Its performance toward catalytic transfer hydrogenation of furfural is comparable to state-of-the-art catalysts and up to four orders higher than other Fe catalysts. Isotopic labeling experiments demonstrate an intermolecular hydride transfer mechanism. First principles simulations, spectroscopic calculations and experiments, and kinetic correlations reveal that the synthesis creates pyrrolic Fe(II)-plN 3 as the active center whose flexibility manifested by being pulled out of the plane, enabled by defects, is crucial for collocating the reagents and allowing the chemistry to proceed. The catalyst catalyzes chemoselectively several substrates and possesses a unique trait whereby the chemistry is hindered for more acidic substrates than the hydrogen donors. This work paves the way toward noble-metal free single-atom catalysts for important chemical reactions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2018

Impact of cross-linking of polymers on transport of salt and water in polyelectrolyte membranes: A mesoscopic simulation study

Our recent atomistic simulation studies demonstrated that the transport properties of salt ions and water in non-crosslinked polymer electrolyte membrane exhibit an intriguing dependence on salt concentration that is opposite to that seen in electrolyte solutions. Here in this paper, we extend our study to probe the influence of the degree of cross-linking of the polymer on the transport properties of salt and water in polymer electrolyte membranes. Towards this objective, we use a coarse-grained model embedded within dissipative particle dynamics (DPD) mesoscale simulations, which allows us to access time scales necessary for studying crosslinked polymer systems. Our DPD simulations on non-crosslinked membranes reproduce results that are in qualitative agreement with our atomistic simulations. For the case of crosslinked membranes, our results demonstrate that the diffusion of salt ions and water is reduced significantly relative to crosslinked systems. However, the trends exhibited by the salt concentration dependence of diffusivities and the coordination of the cations with anions and with the polymer backbone remain qualitatively similar to those observed in non-crosslinked membranes.

36 MATERIALS SCIENCE↗
DOE OSTI2025

Dependent scattering and fractal microstructure determine the transparency of aerogel monoliths

This study reveals how dependent scattering and microstructure significantly affect electromagnetic wave propagation through aerogel monoliths, contributing to their transparency. Light scattering by particle ensembles is considered “dependent” when the scattering properties rely not only on particle size and optical constants but also on their spatial distribution, typically occurring when the average interparticle distance is small in comparison with the wavelength of incident radiation. Addressing dependent scattering requires solving Maxwell’s equations for complex heterogeneous structures, which is computationally demanding and usually limited to sample thicknesses on the same scale as the wavelength. This study combines computer-generated ambigel microstructures of fractal aggregates of polydisperse nanoparticles and the radiative transfer with reciprocal transaction method to predict the transmittance of thick ambigel slabs. Transmittance measurements of ambiently dried aerogel monoliths (ambigels) with porosities from about 50% to 90% closely matched the predicted values for their digital twins. However, ignoring dependent scattering or particle aggregation led to inaccurate predictions. This study validated the computational framework, and its findings offer insights for designing photonic metamaterials and analyzing their interactions with electromagnetic waves.

Yalcin, Refet A. (ORCID:0000000339973494)↗
DOE OSTI2021

Optimizing Error-Bounded Lossy Compression for Scientific Data on GPUs

Error-bounded lossy compression is a critical technique for significantly reducing scientific data volumes. With ever-emerging heterogeneous high-performance computing (HPC) architecture, GPU-accelerated error-bounded compressors (such as CUSZ and cuZFP) have been developed. However, they suffer from either low performance or low compression ratios. To this end, we propose CUSZ+ to target both high compression ratios and throughputs. We identify that data sparsity and data smoothness are key factors for high compression throughputs. Our key contributions in this work are fourfold: (1) We propose an efficient compression workflow to adaptively perform run-length encoding and/or variable-length encoding. (2) We derive Lorenzo reconstruction in decompression as multidimensional partial-sum computation and propose a fine-grained Lorenzo reconstruction algorithm for GPU architectures. (3) We carefully optimize each of CUSZ kernels by leveraging state-of-the-art CUDA parallel primitives. (4) We evaluate CUSZ+ using seven real-world HPC application datasets on V100 and A100 GPUs. Experiments show CUSZ+ improves the compression throughputs and ratios by up to 18.4x and 5.3x, respectively, over CUSZ on the tested datasets.

Tian, Jiannan↗
DOE OSTI2024

IRIS: Exploring Performance Scaling of the Intelligent Runtime System and its Dynamic Scheduling Policies

High-Performance Computing is becoming increasingly heterogeneous, relying on a diverse mix of hardware to achieve good performance. Paradoxically, current drivers and frameworks for these devices typically require separate languages and implementations for each vendor. Furthermore, there are few tools and little support to schedule codes between these devices in a truly heterogeneous manner-partly because of this fragmentation between vendors and the languages each supports. To overcome both limitations, the Intelligent Runtime System (IRIS) was developed. It allows a common task abstraction to automatically be shared among contemporary vendors and is run from a single host-side API. At runtime, IRIS queries the host system and registers which frameworks and drivers are available, these determine which kernels can be used by the scheduler-CPUs via OpenMP, Nvidia GPUs (CUDA), AMD GPUs (HIP), and Intel and Xilinx FPGAs with OpenCL. IRIS enables tasks to be scheduled to any heterogeneous device and resolves to the appropriate kernel binary at runtimeit only uses the devices supported by the system on which it is run. IRIS supports single-task and graph-based expressions of dependencies of tasks. Additionally, IRIS features a range of dynamic scheduling policies, allowing complex chains of tasks and interactions to be executed, relieving the programmer/user from considering the system to assign tasks to devices optimally. This paper presents the peak performance attainable by IRIS over a range of systems-each with different numbers and types of accelerator devices, it highlights the flexibility of IRIS since these devices are truly heterogeneous, relying on different backends (drivers, frameworks, and languages) which historically required unique implementations to utilize them. We then use this peak performance as a baseline to compare increasingly complex chains of tasks (with increasingly complex task dependencies) and evaluate how IRIS copes. Finally, we consider the performance of different IRIS scheduling policies on this range of task graphs.

Johnston, Beau↗
DOE OSTI2024

EchemAMR (electro-chemical microsctructure scale models with adaptive meshing) [SWR-23-111]

A 3D microstructure resolving electrochemical transport and interfacial chemistry solver. Electrode microstructure plays an important role in determining the performance of an electrochemical system, e.g. lithium ion battery. EchemAMR is a microstructure scale model that solves the governing equations for ion transport, electrical current continuity, interfacial chemistry and structural mechanics. Complex microstructure geometries from imaging can be directly imported into EchemAMR. A volume fraction based description of the geometry on Cartesian grid with an immersed interface formulation enables simplified meshing and large-scale simulations with millions of degrees of freedom. EchemAMR has been tested against systems with analytic solutions for numerical convergence and highly resolved lithium ion battery microstructures. EchemAMR demonstrates excellent mass conversation and efficient scaling on heterogenous High-Performance Computing (HPC) with central and graphics processing units.

Sitaraman, Hariswaran↗
DOE OSTI2016

Towards performance-portability of the Albany Finite Element analysis code using the Kokkos library of Trilinos

Performance portability on heterogeneous high-performance computing (HPC) systems is a major challenge faced today by code developers: parallel code needs to execute correctly as well as with high performance on machines with different architectures, operating systems, and software libraries. The Finite Element Method (FEM) is a popular and flexible method for discretizing partial differential equations arising in a wide variety of scientific, engineering, and industry applications that require HPC. This paper presents some preliminary results pertaining to our development of a performance portable implementation of the FEM-based Albany code. Performance portability is achieved using the Kokkos library of Trilinos. We present performance results for two different physics simulations modules in Albany: the Aeras global atmosphere dynamical code and the FELIX land-ice solver. As a result, numerical experiments show that our single code implementation gives reasonable performance across two multi-core/many-core architectures: NVIDIA GPUs and multi-core CPUs.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

Fréchet kernels based on a fractional viscoacoustic wave equation

Incorporating the seismic attenuation into the waveform inversion framework could not only improve the accuracy of the velocity model but also provide an additional Q model. Recently, we proposed a viscoacoustic wave equation assisted by the fractional Laplacian operators to accurately model the wave propagation in heterogeneous attenuating media with computational efficiency. The explicit presence of Q as a coefficient in this equation suggests the potential to conveniently develop its full waveform inversion scheme. In this study, we utilize the adjoint-state method to formulate the computation of the Fréchet kernels with respect to both velocity and attenuation based on this wave equation. These kernels will play a fundamental role in the viscoacoustic multiparameter waveform inversion.

Xing, Guangchi↗
DOE OSTI2021

Rapid simulations of hyperspectral near-field images of three-dimensional heterogeneous surfaces

The scattering-type scanning near-field optical microscope (s-SNOM) has emerged as a powerful tool for resolving nanoscale inhomogeneities in laterally heterogeneous samples. However, most analytical models used to predict the scattering near-field signals are assuming homogenous landscapes (bulk materials), resulting in inconsistencies when applied to samples with more complex configurations. In this work, we combine the point-dipole model (PDM) to the finite-element method (FEM) to account for the lateral and vertical heterogeneities while keeping the computation time manageable. Full images, spectra, or hyperspectral line profiles can be simulated by calculating the self-consistent dipole radiation demodulated at higher harmonics of the tip oscillation, mimicking real experimental procedures. Using this formalism, we clarify several important yet puzzling experimental observations in near-field images on samples with rich typography and complex material compositions, heterostructures of two-dimensional material flakes, and plasmonic antennas. The developed method serves as a basis for future investigations of nano-systems with nontrivial topography.

36 MATERIALS SCIENCE↗
DOE OSTI2015

Controlling Subsurface Fractures and Fluid Flow: A Basic Research Agenda

From beneath the surface of the earth, we currently obtain about 80-percent of the energy our nation consumes each year. In the future we have the potential to generate billions of watts of electrical power from clean, green, geothermal energy sources. Our planet’s subsurface can also serve as a reservoir for storing energy produced from intermittent sources such as wind and solar, and it could provide safe, long-term storage of excess carbon dioxide, energy waste products and other hazardous materials. However, it is impossible to underestimate the complexities of the subsurface world. These complexities challenge our ability to acquire the scientific knowledge needed for the efficient and safe exploitation of its resources. To more effectively harness subsurface resources while mitigating the impacts of developing and using these resources, the U.S. Department of Energy established SubTER – the Subsurface Technology and Engineering RD&D Crosscut team. This DOE multi-office team engaged scientists and engineers from the national laboratories to assess and make recommendations for improving energy-related subsurface engineering. The SubTER team produced a plan with the overall objective of “adaptive control of subsurface fractures and fluid flow.”This plan revolved around four core technological pillars—Intelligent Wellbore Systems that sustain the integrity of the wellbore environment; Subsurface Stress and Induced Seismicity programs that guide and optimize sustainable energy strategies while reducing the risks associated with subsurface injections; Permeability Manipulation studies that improve methods of enhancing, impeding and eliminating fluid flow; and New Subsurface Signals that transform our ability to see into and characterize subsurface systems. The SubTER team developed an extensive R&D plan for advancing technologies within these four core pillars and also identified several areas where new technologies would require additional basic research. In response, the Office of Science, through its Office of Basic Energy Science (BES), convened a roundtable consisting of 15 national lab, university and industry geoscience experts to brainstorm basic research areas that underpin the SubTER goals but are currently underrepresented in the BES research portfolio. Held in Germantown, Maryland on May 22, 2015, the round-table participants developed a basic research agenda that is detailed in this report. Highlights include the following: -A grand challenge calling for advanced imaging of stress and geological processes to help understand how stresses and chemical substances are distributed in the subsurface—knowledge that is critical to all aspects of subsurface engineering; -A priority research direction aimed at achieving control of fluid flow through fractured media; -A priority research direction aimed at better understanding how mechanical and geochemical perturbations to subsurface rock systems are coupled through fluid and mineral interactions; -A priority research direction aimed at studying the structure, permeability, reactivity and other properties of nanoporous rocks, like shale, which have become critical energy materials and exhibit important hallmarks of mesoscale materials; -A cross-cutting theme that would accelerate development of advanced computational methods to describe heterogeneous time-dependent geologic systems that could, among other potential benefits, provide new and vastly improved models of hydraulic fracturing and its environmental impacts; -A cross-cutting theme that would lead to the creation of “geo-architected materials” with controlled repeatable heterogeneity and structure that can be tested under a variety of thermal, hydraulic, chemical and mechanical conditions relevant to subsurface systems; -A cross-cutting theme calling for new laboratory studies on both natural and geo-architected subsurface materials that deploy advanced high-resolution 3D imaging and chemical analysis methods to determine the ;rates and mechanisms of fluid-rock processes, and to test predictive models of such phenomena. Many of the key energy challenges of the future demand a greater understanding of the subsurface world in all of its complexity. This greater under- standing will improve the ability to control and manipulate the subsurface world in ways that will benefit both the economy and the environment. This report provides specific basic research pathways to address some of the most fundamental issues of energy-related subsurface engineering.

15 GEOTHERMAL ENERGY↗
DOE OSTI2017

FY17 ASC P&EM L2 Milestone 6009: Demonstrate Thread Scalability within Aria on Both Sides of Trinity

The use of next-generation platforms (NGPs), also known as advanced technology systems (ATS), that incorporate many-core and heterogeneous architectures for scientific computing represent a tectonic shift in computing hardware design that will require massive development work within the Sierra applications to harness these systems to their full potential. The completion of this milestone represents a first step towards this effort by threading many of the computational kernels within Sierra/Aria.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

An energy-based coupling approach to nonlocal interface problems.

Nonlocal models provide accurate representations of physical phenomena ranging from fracture mechanics to complex subsurface flows, settings in which traditional partial differential equation models fail to capture effects caused by long-range forces at the microscale and mesoscale. However, the application of nonlocal models to problems involving interfaces, such as multimaterial simulations and fluid-structure interaction, is hampered by the lack of a physically consistent interface theory which is needed to support numerical developments and, among other features, reduces to classical models in the limit as the extent of nonlocal interactions vanish. In this paper, we use an energy-based approach to develop a formulation of a nonlocal interface problem which provides a physically consistent extension of the classical perfect interface formulation for partial differential equations. Numerical examples in one and two dimensions validate the proposed framework and demonstrate the scope of our theory.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2026

Rapid Gamma Simulations of TRISO Fuel Elements

As energy demand rises, nuclear energy, particularly from reactors that use tristructural isotropic (TRISO) fuels, has gained attention due to the fuel’s enhanced resistance to radiation damage and high temperatures. This report investigates the modeling capabilities of the Gamma Detector Response and Analysis Software (GADRAS) for TRISO fuels, focusing on the gamma signatures of TRISO particles, which have not been extensively explored. Using the Monte Carlo N-Particle (MCNP) code as a benchmark, we developed both homogeneous and heterogeneous models of TRISO pebbles to analyze gamma spectra. Our findings reveal that the homogeneous and heterogeneous models produced different gamma signatures. Additionally, the GADRAS heterogeneous model significantly reduces computation times compared to MCNP, enabling effective modeling of gamma signatures for safeguards applications. This advancement is essential for the International Atomic Energy Agency (IAEA) in detecting anomalies and potential smuggling attempts in TRISO reactor fuel elements.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2023

Efficient solutions of fermionic systems using artificial neural networks

In this study, we explore the similarities and differences between variational Monte Carlo techniques that employ conventional and artificial neural network representations of the ground-state wave function for fermionic systems. Our primary focus is on shallow neural network architectures, specifically the restricted Boltzmann machine, and we examine unsupervised learning algorithms that are appropriate for modeling complex many-body correlations. We assess the advantages and drawbacks of conventional and neural network wave functions by applying them to a range of circular quantum dot systems. Our findings, which include results for systems containing up to 90 electrons, emphasize the efficient implementation of these methods on both homogeneous and heterogeneous high-performance computing facilities.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1988

Telescience, an operational approach to science investigation

The NASA Science and Applications Information System, which is based on telescience and must provide remote interaction between information system services in space and on the ground, is discussed. An infrastructure of networked facilities and institutionally provided support services is being developed. The technologies involved with providing telescience capability are examined, including automated data management services, new data acquisition systems, user support environment for system access, and the capability to access heterogeneous data bases and computational facilities from remote locations.

Weiss, James R.↗
NASA NTRSDate not supplied

Enabling Execution of a Legacy CFD Mini Application on Accelerators Using OpenMP

We describe the process and outcome of our efforts to port a legacy Fortran benchmark code to heterogeneous GPU-accelerated computing architectures using OpenMP. The benchmark code is one of the multi-zone NAS Parallel Benchmarks (NPB-MZ) called SP-MZ. This “mini-app” mimics the computation and data movement that is found in popular legacy and modern implicit computational fluid dynamics (CFD)solvers. Our objective was to examine how efficiently legacy Fortran codes can be ported to accelerators by leveraging OpenMP directives. We describe the development and optimization process and demonstrate the performance impact of various code modifications. We show select profiling results from the Nvidia nvvp profiler to help others diagnose and overcome performance issues in their own applications. We present results for two compute systems endowed with Nvidia V100 accelerators.

Ioannis Nompelis↗
NASA NTRSDate not supplied

Benchmarking and Performance of the NASA Multiscale Analysis Tool

The NASA Multiscale Analysis Tool (NASMAT) is as a “plug and play,” software package which utilizes multiscale recursive micromechanics as a platform for massively multiscale modeling of hierarchical materials and structures subjected to thermomechanical. This paper is intended to give an overview of the design of NASMAT and how the design supports modularity, upgradability and maintainability, interoperability, and utility. First, the software architecture and hierarchy will be explored. Details on each of the 11 NASMAT procedures and the arrangement of NASMAT data will be presented. Application program interfaces (APIs) that were developed to facilitate the communication of NASMAT with other programs will be described. The intended application for NASMAT is massively multiscale modeling on high performance computing systems. As such, results benchmarking the performance of the integration of NASMAT with the Abaqus commercial finite element method software are also presented.

Multiscale Modeling↗