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.

SourceResearch connectionFrom yearThrough year

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

At least 577 records · Page 32

All sources NASA NTRS DOE OSTI

DOE OSTI2014

Scale-Dependent Fracture-Matrix Interactions And Their Impact on Radionuclide Transport - Final Report

Matrix diffusion and adsorption within a rock matrix are widely regarded as important mechanisms for retarding the transport of radionuclides and other solutes in fractured rock (e.g., Neretnieks, 1980; Tang et al., 1981; Maloszewski and Zuber, 1985; Novakowski and Lapcevic, 1994; Jardine et al., 1999; Zhou and Xie, 2003; Reimus et al., 2003a,b). When remediation options are being evaluated for old sources of contamination, where a large fraction of contaminants reside within the rock matrix, slow diffusion out of the matrix greatly increases the difficulty and timeframe of remediation. Estimating the rates of solute exchange between fractures and the adjacent rock matrix is a critical factor in quantifying immobilization and/or remobilization of DOE-relevant contaminants within the subsurface. In principle, the most rigorous approach to modeling solute transport with fracture-matrix interaction would be based on local-scale coupled advection-diffusion/dispersion equations for the rock matrix and in discrete fractures that comprise the fracture network (Discrete Fracture Network and Matrix approach, hereinafter referred to as DFNM approach), fully resolving aperture variability in fractures and matrix property heterogeneity. However, such approaches are computationally demanding, and thus, many predictive models rely upon simplified models. These models typically idealize fracture rock masses as a single fracture or system of parallel fractures interacting with slabs of porous matrix or as a mobile-immobile or multi-rate mass transfer system. These idealizations provide tractable approaches for interpreting tracer tests and predicting contaminant mobility, but rely upon a fitted effective matrix diffusivity or mass-transfer coefficients. However, because these fitted parameters are based upon simplified conceptual models, their effectiveness at predicting long-term transport processes remains uncertain. Evidence of scale dependence of effective matrix diffusion coefficients obtained from tracer tests highlights this point and suggests that the underlying mechanisms and relationship between rock and fracture properties are not fully understood in large complex fracture networks. In this project, we developed a high-resolution DFN model of solute transport in fracture networks to explore and quantify the mechanisms that control transport in complex fracture networks and how these may give rise to observed scale-dependent matrix diffusion coefficients. Results demonstrate that small scale heterogeneity in the flow field caused by local aperture variability within individual fractures can lead to long-tailed breakthrough curves indicative of matrix diffusion, even in the absence of interactions with the fracture matrix. Furthermore, the temporal and spatial scale dependence of these processes highlights the inability of short-term tracer tests to estimate transport parameters that will control long-term fate and transport of contaminants in fractured aquifers.

58 GEOSCIENCES↗

DOE OSTI2026

The ocean model for E3SM global applications: Omega version 0.1.0 – a new high-performance computing code for exascale architectures

This paper introduces Omega, the Ocean Model for E3SM Global Applications. Omega is a new ocean model designed to run efficiently on high performance computing (HPC) platforms, including exascale heterogeneous architectures with accelerators, such as Graphics Processing Units (GPUs). Omega is written in C and uses the Kokkos performance portability library. These were chosen because they are well-supported and will help future-proof Omega for upcoming HPC architectures. Omega will eventually replace the Model for Prediction Across Scales-Ocean (MPAS-Ocean) in the US Department of Energy's (DOE's) Energy Exascale Earth System Model (E3SM). Omega runs on unstructured horizontal meshes with variable-resolution capability and implements the same horizontal discretization as MPAS-Ocean. This work documents the design and performance of Omega Version 0.1.0 (Omega-V0), which solves the shallow water equations with passive tracers and is the first step towards the full primitive equation ocean model. On Central Processing Units (CPUs), Omega-V0 is 1.4 times faster than MPAS-Ocean with the same configuration. Omega-V0 is more efficient on GPUs than CPUs on a per-watt basis – by a factor of 5.3 on Frontier and 3.6 on Aurora, two of the world's fastest exascale computers.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2023

Building the I (Interoperability) of FAIR for performance reproducibility of large-scale composable workflows in RECUP

Abstract-Scientific computing communities increasingly run their experiments using complex data- and compute-intensive workflows that utilize distributed and heterogeneous architectures targeting numerical simulations and machine learning, often executed on the Department of Energy Leadership Computing Facilities (LCFs). We argue that a principled, systematic approach to implementing FAIR principles at scale, including fine-grained metadata extraction and organization, can help with the numerous challenges to performance reproducibility posed by such workflows. We extract workflow patterns, propose a set of tools to manage the entire life cycle of performance metadata, and aggregate them in an HPC-ready framework for reproducibility (RECUP). We describe the challenges in making these tools interoperable, preliminary work, and lessons learned from this experiment.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2023

HetArch: Heterogeneous Microarchitectures for Superconducting Quantum Systems

Noisy Intermediate-Scale Quantum Computing (NISQ) has dominated headlines in recent years, with the longer-term vision of Fault-Tolerant Quantum Computation (FTQC) offering significant potential but at currently intractable resource costs and quantum error correction (QEC) overheads. For problems of interest, FTQC will require millions of physical qubits with long coherence times, high-fidelity gates, and compact sizes to surpass classical systems. Just as heterogeneous specialization has offered scaling benefits in classical computing, it is likewise gaining interest in FTQC. However, systematic use of heterogeneity in either hardware or software elements of FTQC systems remains a serious challenge due to the vast design space and the variable physical constraints. This paper meets the challenge of making heterogeneous FTQC design practical by introducing HetArch, a toolbox for designing heterogeneous quantum systems, and using it to explore heterogeneous design scenarios. Using a hierarchical approach, we successively break quantum algorithms into smaller operations (akin to classical application kernels), thus greatly simplifying the design space and resulting tradeoffs. Specializing to superconducting systems, we then design optimized heterogeneous hardware composed of varied superconducting devices, abstracting physical constraints into design rules that enable devices to be assembled into standard cells optimized for specific operations, which, in turn, form heterogeneous modules optimized for quantum subroutines. Finally, we provide a heterogeneous design space exploration framework which reduces the simulation burden by a factor of 10^4 or more and allows us to characterize optimal design points. We use these techniques to design superconducting quantum modules for entanglement distillation, error correction, and code teleportation, reducing error rates by 2.6×, 10.7×, and 3.4× compared to homogeneous systems.

Quantum Computing, Quantum Physics, Computer Archi↗

DOE OSTI2018

Energy Efficient Formaldehyde Synthesis by Direct Hydrogenation of Carbon Monoxide in Functionalized Metal–Organic Frameworks

Formaldehyde (CH 2 O) is a highly versatile platform chemical with an annual production of over 30 megatons. The current most widely used industrial method for formaldehyde synthesis is very energy intensive and plagued with significant exergy losses. We propose a highly efficient heterogeneous nanoporous catalyst capable of direct hydrogenation of CO to CH 2 O followed by condensation of CH 2 O from a mixture of CO and H 2 , which are non-condensable gases. The key to this synthesis is a Lewis Pair (LP) functionalized metal-organic framework (MOF) heterogeneous porous catalyst. We have computed reaction pathways, barriers, and uncertainties for CO hydrogenation on a LP functionalized MOF using various density functional theory (DFT) methods and high-level ab initio methods. We have assessed the uncertainties in the energetics inherent in the DFT functionals. Here, we predict that the thermodynamics of CH 2 O synthesis will be enhanced relative to gas phase synthesis and more importantly, will not be equilibrium limited because the barrier to product desorption from the MOF is much lower than the barrier for the reverse reaction. Hence, the product will be continuously removed from the process, facilitating very high overall conversion rates. Our proposed process will dramatically improve the sustainability of CH 2 O synthesis.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2018

Effect of Cell-Boundary Currents on the Fuel Burnup Characteristics of VVER-1000 Light-Water Reactor

The results of a computational study of the effect of nonzero currents at FA boundaries on the characteristics of fuel burnup in VVER-1000 are presented. A variant with 1/3 mixed fuel loaded into the core was studied. The TRIFON 2.1 code was used to calculate the time dependence of the characteristics of FA with uranium and mixed fuel and the SHERHAN code was used for computational modeling of three-dimensional burnup; the codes were modified on the basis of an improved heterogeneous method. It is shown by a computational method that in VVER-1000 with a mixed fuel load nonzero currents at FA boundaries, matched with their conditions inside the reactor, give a moderate effect in the multiplication coefficient – ≤0.2–0.35%, in the coefficient of nonuniformity of the power distribution – ≤1.5%, and in the burnup of uranium and mixed fuel in FA over a fuel cycle – ≤0.6%.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

DOE OSTI2023

SuperNeuro: A Fast and Scalable Simulator for Neuromorphic Computing

In many neuromorphic workflows, simulators play a vital role for important tasks such as training spiking neural networks, running neuroscience simulations, and designing, implementing, and testing neuromorphic algorithms. Currently available simulators cater to either neuroscience workflows (e.g., NEST and Brian2) or deep learning workflows (e.g., BindsNET). Problematically, the neuroscience-based simulators are slow and not very scalable, and the deep learning-based simulators do not support certain functionalities that are typical of neuromorphic workloads (e.g., synaptic delay). In this paper, we address this gap in the literature and present SuperNeuro, which is a fast and scalable simulator for neuromorphic computing capable of both homogeneous and heterogeneous simulations as well as GPU acceleration. We also present preliminary results that compare SuperNeuro to widely used neuromorphic simulators such as NEST, Brian2, and BindsNET in terms of computation times. We demonstrate that SuperNeuro can be approximately 10×--300× faster than some of the other simulators for small sparse networks. On large sparse and large dense networks, SuperNeuro can be approximately 2.2×--3.4× faster than the other simulators, respectively.

Date, Prasanna↗

DOE OSTI2016

Heterogeneous Reduction Pathways for Hg(II) Species on Dry Aerosols: A First-Principles Computational Study

Here, the atmospheric lifetime of mercury is greatly impacted by redox chemistry resulting from the high deposition rate of reactive mercury (Hg(II)) compared to elemental mercury (Hg 0 ). Recent laboratory and field studies have observed the reduction of Hg(II) but the chemical mechanism for this reaction has not been identified. Recent laboratory studies have shown that the reduction reaction is heterogeneous and can occur on iron and sodium chloride aerosol surfaces. This study explores the use of density functional theory calculations to discern the reduction pathways of HgCl 2 , HgBr 2 , Hg(NO 3 ) 2 , and HgSO 4 on clean Fe(110), NaCl(100), and NaCl(111) Na surfaces. In doing so, potential energy surfaces have been prepared for the various reduction pathways, indicating that the reduction pathway leading to the production of gas-phase elemental mercury is highly favorable on Fe(110) and NaCl(111) Na . Moreover, the Fe(110) surface requires an external energy source of approximately 0.5 eV to desorb the reduced mercury, whereas the NaCl(111) Na surface requires no energy input. The results indicate that a number of mercury species can be reduced on metallic iron and sodium chloride surfaces, which are known aerosol components, and that a photochemical reaction involving the aerosol surface is likely needed for the reaction to be catalytic.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2019

Application of the ESMACS Binding Free Energy Protocol to a Multi-Binding Site Lactate Dehydogenase A Ligand Dataset

Over the past two decades, the use of fragment-based lead generation has become a common, mature approach to identify tractable starting points in chemical space for the drug discovery process. This approach naturally involves the study of the binding properties of highly heterogeneous ligands. Such datasets challenge computational techniques to provide comparable binding free energy estimates from different binding modes. The performance of a range of statistically robust ensemble-based binding free energy calculation protocols, called ESMACS (enhanced sampling of molecular dynamics with approximation of continuum solvent), is evaluated. Ligands designed to target two binding pockets in the lactate dehydogenase, a target protein, which vary in size, charge, and binding mode, are studied. When compared to experimental results, excellent statistical rankings are obtained across this highly diverse set of ligands. In addition, three approaches to account for entropic contributions are investigated: 1) normal mode analysis, 2) weighted solvent accessible surface area (WSAS), and 3) variational entropy. Normal mode analysis and WSAS correlate strongly with each other—although the latter is computationally far cheaper—but do not improve rankings. Variational entropy corrects exaggerated discrimination of ligands bound in different pockets but creates three outliers which reduce the quality of the overall ranking.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2017

Tandem Diels-Alder Reaction of Dimethylfuran and Ethylene and Dehydration to para -Xylene Catalyzed by Zeotypic Lewis Acids

The zeotypic Lewis acids Sn-BEA, Zr-BEA, and Ti-BEA have recently been discussed to catalyze the synthesis of p-xylene by dehydrative aromatization of the Diels–Alder product between 2,5-dimethylfuran and ethylene. While it has been shown that these Lewis acids can catalyze the dehydration of the Diels–Alder cycloadduct, the tandem scheme precludes decoupling of the two steps needed to infer whether these same catalysts can catalyze the Diels–Alder step. We have employed electronic structure calculations and microkinetic modelling to investigate the Diels–Alder aromatization of 2,5-dimethylfuran and ethylene to p-xylene over the Lewis-acidic zeotypes Sn-, Zr-, and Ti-BEA. We show that there is only minor catalysis of the Diels–Alder reaction, solely attributable to confinement phenomena varying with the translational freedom allowed to the species inside the zeolite. Microkinetic modelling and sensitivity analysis of the computed rates show that the heterogeneous Diels–Alder pathway does not contribute to the overall rate, and that the homogeneous cycloaddition is rate-limiting at high acid site concentrations. Only the partially hydrolyzed (“open”) Lewis acid sites are found to be catalytically active, with moderately Brønsted-acidic silanol groups formed, which catalyze 2,5-dimethylfuran hydrolysis. Of the Lewis acids tested in this work, Zr-BEA and Sn-BEA have similar activities, in agreement with experiment, whereas Ti-BEA is found to be inactive, indicating that the recently reported Ti-BEA activity was likely a result of Brønsted-acidic defect sites.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

DOE OSTI2024

Recent Advances in X-ray Absorption Near Edge Structure (XANES) Simulation for Catalysis: Theories and Applications

X-ray absorption spectroscopy (XAS) is a powerful spectroscopic technique for characterizing catalytic active sites. In particular, X-ray Absorption Near Edge Structure (XANES) provides insights into the oxidation state and local coordination environment of the absorbing atom, shedding light on the catalytic structure. Computational XANES simulations provide one-on-one correspondence between the molecular structure and their corresponding spectra, making them an effective approach for interpreting experimental XANES features. This work outlines various theories used to simulate XANES spectra and highlights recent advancements in computational XANES applications for homogeneous, heterogeneous, and single-atom catalysts focusing on identifying the active site structures, the site heterogeneity, and the dynamic evolution of catalysts.

Xu, Jiayi↗

DOE OSTI2012

Potential for Improved Intelligence Quotient Using Volumetric Modulated Arc Therapy Compared With Conventional 3-Dimensional Conformal Radiation for Whole-Ventricular Radiation in Children

Purpose: To compare volumetric modulated arc therapy (VMAT) with 3-dimensional conformal radiation therapy (3D-CRT) in the treatment of localized intracranial germinoma. We modeled the effect of the dosimetric differences on intelligence quotient (IQ). Method and Materials: Ten children with intracranial germinomas were used for planning. The prescription doses were 23.4 Gy to the ventricles followed by 21.6 Gy to the tumor located in the pineal region. For each child, a 3D-CRT and full arc VMAT was generated. Coverage of the target was assessed by computing a conformity index and heterogeneity index. We also generated VMAT plans with explicit temporal lobe sparing and with smaller ventricular margin expansions. Mean dose to the temporal lobe was used to estimate IQ 5 years after completion of radiation, using a patient age of 10 years. Results: Compared with the 3D-CRT plan, VMAT improved conformality (conformity index 1.10 vs 1.85), with slightly higher heterogeneity (heterogeneity index 1.09 vs 1.06). The averaged mean doses for left and right temporal lobes were 31.3 and 31.7 Gy, respectively, for VMAT plans and 37.7 and 37.6 Gy for 3D-CRT plans. This difference in mean temporal lobe dose resulted in an estimated IQ difference of 3.1 points at 5 years after radiation therapy. When the temporal lobes were explicitly included in the VMAT optimization, the mean temporal lobe dose was reduced 5.6-5.7 Gy, resulting in an estimated IQ difference of an additional 3 points. Reducing the ventricular margin from 1.5 cm to 0.5 cm decreased mean temporal lobe dose 11.4-13.1 Gy, corresponding to an estimated increase in IQ of 7 points. Conclusion: For treatment of children with intracranial pure germinomas, VMAT compared with 3D-CRT provides increased conformality and reduces doses to normal tissue. This may result in improvements in IQ in these children.

62 RADIOLOGY AND NUCLEAR MEDICINE↗

DOE OSTI2019

Water follows polar and nonpolar protein surface domains

The conformation of water around proteins is of paramount importance, as it determines protein interactions. Although the average water properties around the surface of proteins have been provided experimentally and computationally, protein surfaces are highly heterogeneous. Therefore, it is crucial to determine the correlations of water to the local distributions of polar and nonpolar protein surface domains to understand functions such as aggregation, mutations, and delivery. By using atomistic simulations, we investigate the orientation and dynamics of water molecules next to 4 types of protein surface domains: negatively charged, positively charged, and charge-neutral polar and nonpolar amino acids. The negatively charged amino acids orient around 98% of the neighboring water dipoles toward the protein surface, and such correlation persists up to around 16 Å from the protein surface. The positively charged amino acids orient around 94% of the nearest water dipoles against the protein surface, and the correlation persists up to around 12 Å. The charge-neutral polar and nonpolar amino acids are also orienting the water neighbors in a quantitatively weaker manner. A similar trend was observed in the residence time of the nearest water neighbors. These findings hold true for 3 technically important enzymes (PETase, cytochrome P450, and organophosphorus hydrolase). Our results demonstrate that the water−amino acid degree of correlation follows the same trend as the amino acid contribution in proteins solubility, namely, the negatively charged amino acids are the most beneficial for protein solubility, then the positively charged amino acids, and finally the charge-neutral amino acids.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2012

Effect of Nonlinearity in Hybrid Kinetic Monte Carlo-Continuum Models

Recently there has been interest in developing efficient ways to model heterogeneous surface reactions with hybrid computational models that couple a KMC model for a surface to a finite difference model for bulk diffusion in a continuous domain. We consider two representative problems that validate a hybrid method and also show that this method captures the combined effects of nonlinearity and stochasticity. We first validate a simple deposition/dissolution model with a linear rate showing that the KMC-continuum hybrid agrees with both a fully deterministic model and its analytical solution. We then study a deposition/dissolution model including competitive adsorption, which leads to a nonlinear rate, and show that, in this case, the KMC-continuum hybrid and fully deterministic simulations do not agree. However, we are able to identify the difference as a natural result of the stochasticity coming from the KMC surface process. Because KMC captures inherent fluctuations, we consider it to be more realistic than a purely deterministic model. Therefore, we consider the KMC-continuum hybrid to be more representative of a real system.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

DOE OSTI2022

High Performance Adaptive Physics Refinement to Enable Large-Scale Tracking of Cancer Cell Trajectory

The ability to track simulated cancer cells through the circulatory system, important for developing a mechanistic understanding of metastatic spread, pushes the limits of today's supercomputers by requiring the simulation of large fluid volumes at cellular-scale resolution. To overcome this challenge, we introduce a new adaptive physics refinement (APR) method that captures cellular-scale interaction across large domains and leverages a hybrid CPU-GPU approach to maximize performance. Through algorithmic advances that integrate multi-physics and multi-resolution models, we establish a finely resolved window with explicitly modeled cells coupled to a coarsely resolved bulk fluid domain. In this work we present multiple validations of the APR framework by comparing against fully resolved fluid-structure interaction methods and employ techniques, such as latency hiding and maximizing memory bandwidth, to effectively utilize heterogeneous node architectures. Collectively, these computational developments and performance optimizations provide a robust and scalable framework to enable system-level simulations of cancer cell transport.

59 BASIC BIOLOGICAL SCIENCES↗

DOE OSTI2026

SoS-SDQN: System of Systems Software-defined Quantum Networking

Quantum networks are needed for quantum domain applications that may run across different network deployments - point-to-point, multi-node networks, and possibly across inter-domain quantum networks, like the quantum internet. Unlike classical computing networks, quantum networks involve heterogeneous systems nodes that currently require local and manual control. This needs a unified control approach to help them integrate, work seamlessly, and have global knowledge. Software-defined Networking (SDN) has been successfully leveraged in classical networking for seamless and software-driven control of network infrastructure and packet switching, but not in management of quantum network applications across heterogeneous systems. In this paper, we review the current state of the art and present early developments of a System-of-Systems Software-defined Quantum Networking architecture (SoS-SDQN), a generic architecture that supports software-driven quantum network experiments across heterogeneous quantum systems. The architecture modifies the generic SDN to address the domain requirements of quantum networks and proposes a multilevel SDQN to provide a global view of network status, running applications, and control of incorporated heterogeneous quantum systems. We also design and implement a SoS SouthBound Quantum Interface (SoS-SBQI), a quantum infrastructure protocol that abstracts automation of applications across the network.

Alnajjar, Anees [ORNL] (ORCID:0000000237101601)↗

DOE OSTI2023

Fierro Version 2.x

FIERRO is a parallel C++ code designed to simulate fluid mechanics, heat transfer, and solid mechanics in two- and three-dimensional space. FIERRO is written to run on homogeneous (CPU) and heterogeneous (CPU+GPU) high performance computing machines. Fierro can aid a) modeling and design efforts that have historically relied on commercial implicit and explicit finite element codes, b) numerical methods research, c) manufacturing research, and d) computer science research. The code contains diverse numerical methods to solve the governing physics equations for both quasi-static and dynamic problems. Mathematical optimization solvers are coupled to the numerical methods to research topology and shape optimization that has application to additive manufacturing, and to create novel numerical approaches. Phase-field methods with micromechanical solvers are provided to simulate microstructure formation and evolution in manufacturing processes. The micromechanical solvers can also help research efforts create continuum-scale constitutive models for solids, as a function of the microstructure, in situ in a calculation or in a stand-alone manner. No physical data exists within the code.

Morgan, Nathaniel↗

DOE OSTI2023

Fierro

FIERRO is a parallel C++ code designed to simulate fluid mechanics, heat transfer, and solid mechanics in two- and three dimensional space. FIERRO is written to run on homogeneous (CPU) and heterogeneous (CPU+GPU) high performance computing machines. Fierro can aid a) modeling and design efforts that have historically relied on commercial implicit and explicit finite element codes, b) numerical methods research, c) manufacturing research, and d) computer science research. The code contains diverse numerical methods to solve the governing physics equations for both quasi-static and dynamic problems. Mathematical optimization solvers are coupled to the numerical methods to research topology and shape optimization that has application to additive manufacturing, and to create novel numerical approaches. Phase-field methods with micromechanical solvers are provided to simulate microstructure formation and evolution in manufacturing processes. The micromechanical solvers can also help research efforts create continuum-scale constitutive models for solids, as a function of the microstructure, in situ in a calculation or in a stand-alone manner. No physical data exists within the code.

Morgan, Nathaniel↗