Search NASA⌕ Search

SEARCH · Search NASA

Results for “Zero point energy”

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
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

Spin crossovers and superdiffusion in the one-dimensional Hubbard model

We use tools from integrability and generalized hydrodynamics to study finite-temperature dynamics in the one-dimensional Hubbard model. First, we examine charge, spin, and energy transport away from half-filling and zero magnetization, focusing on the strong coupling regime where we identify a rich interplay of temperature and energy scales, with crossovers between distinct dynamical regimes. We identify an intermediate-temperature regime analogous to the spin-incoherent Luttinger liquid, where spin degrees of freedom are hot but charge degrees of freedom are at low temperature. We demonstrate that the spin Drude weight exhibits sharp features at the crossover between this regime and the low-temperature Luttinger liquid regime, which are absent in the charge and energy response, and rationalize this behavior in terms of the properties of Bethe ansatz quasiparticles.We then turn to the dynamics along special lines in the phase diagram corresponding to half-filling and/or zero magnetization where on general grounds we anticipate that the transport is subballistic but superdiffusive. We provide analytical and numerical evidence for Kardar-Parisi-Zhang (KPZ) dynamical scaling (with length and time scales related via x ~ t 2/3 ) along both lines and at the SO(4)-symmetric point where they intersect. Furthermore, our results suggest that both spin-coherence crossovers and KPZ scaling may be accessed in near-term experiments with optical lattice Hubbard emulators.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2021

PyMTensor 1.0.0

SAND2021-1654 O The Python Material Tensor code (PyMTensor) applies crystal symmetries to arbitrary material tensors. It uses symbolic algebra to exactly determine the zero tensor components and the relationships between the nonzero components for all crystallographic point groups. It is capable of computing components of higher-order material tensors needed when treating nonlinear effects. It has a simple and flexible interface for inputting tensor rank information and any known symmetries between tensor indices.Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Jensen, DanielS↗
DOE OSTI2022

A Novel Grid-forming Voltage Control Strategy for Supplying Unbalanced Microgrid Loads Using Inverter-based Resources

This paper presents a novel grid-forming voltage control strategy for a battery energy storage system to maintain balanced three-phase output voltages when serving unbalanced loads. A stationary reference frame (αβ) based control scheme is proposed to regulate positive-sequence and negative-sequence voltages. Compared with the conventional rotating reference frame (dq) based control scheme, the proposed scheme shows better dynamic performance. Then, we analyze the system zero-sequence network and propose adding a grounding transformer to the Y/Yg connected output transformer to further reduce the voltage unbalance at the point of common coupling. The simulation results demonstrate the effective performance of the proposed voltage control scheme. Based on the results, the power-voltage unbalance curve is derived for different output transformer configurations to establish the relationship between the power unbalance limit and the voltage unbalance limit for microgrid power scheduling.

Xu, Bei↗
DOE OSTI2023

Probing the chiral magnetic wave with charge-dependent flow measurements in Pb-Pb collisions at the LHC

The Chiral Magnetic Wave (CMW) phenomenon is essential to provide insights into the strong interaction in QCD, the properties of the quark-gluon plasma, and the topological characteristics of the early universe, offering a deeper understanding of fundamental physics in high-energy collisions. Measurements of the charge-dependent anisotropic flow coefficients are studied in Pb-Pb collisions at center-of-mass energy per nucleon-nucleon collision $\sqrt{s_{\textrm{NN}}}$ = 5.02 TeV to probe the CMW. In particular, the slope of the normalized difference in elliptic (v 2 ) and triangular (v 3 ) flow coefficients of positively and negatively charged particles as a function of their event-wise normalized number difference, is reported for inclusive and identified particles. The slope ${r}_3^{\textrm{Norm}}$ is found to be larger than zero and to have a magnitude similar to ${r}_2^{\textrm{Norm}}$, thus pointing to a large background contribution for these measurements. Furthermore, ${r}_2^{\textrm{Norm}}$ can be described by a blast wave model calculation that incorporates local charge conservation. In addition, using the event shape engineering technique yields a fraction of CMW (f CMW ) contribution to this measurement which is compatible with zero. This measurement provides the very first upper limit for f CMW , and in the 10–60% centrality interval it is found to be 26% (38%) at 95% (99.7%) confidence level.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2022

Extending a scaling equation of state to QCD

Whether quantum chromodynamics (QCD) exhibits a phase transition at finite temperature and density is an open question. It is important for hydrodynamic modeling of heavy ion collisions and neutron-star mergers. Lattice QCD simulations have definitively shown that the transition from hadrons to quarks and gluons is a crossover when the baryon chemical potential is zero or small. Here, we combine the parametric scaling equation of state, usually associated with the three-dimensional Ising model, with a background equation of state based on a smooth crossover from hadrons to quarks and gluons. Comparison to experimental data from the Beam Energy Scan II at the Relativistic Heavy Ion Collider or in heavy ion experiments at other accelerators may allow the critical exponents and amplitudes in the scaling equation of state to be determined for QCD if a critical point exists.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2025

Performance Evaluation of Intelligent Solar Control Software Through Hardware-in-the-Loop (CRADA Final Report)

Recent research has highlighted the potential for solar to act as a zero-marginal-cost and zero-emission flexibility resource on the bulk power system when operated with advanced control systems. To increase the performance of these systems, leading technologies, including machine learning (ML) and hierarchical inverter set point allocation, have been developed by Latimer Controls, Inc. to estimate the headroom of large PV plants for grid operation and control; however, these technologies lack comprehensive validation under real-world application scenarios. Latimer Controls, Inc. received two voucher awards for research at a national laboratory from the Department of Energy American Made Solar Prize Round 6. The National Renewable Energy Laboratory (NREL) was selected to collaborate with Latimer staff to conduct a performance evaluation of Latimer PV control software. The NREL team will develop a hardware-in-the-loop (HIL) testbed to perform testing and validation of the Latimer PV control technology in a de-risked yet realistic testbed environment. Latimer and NREL worked together to analyze the test data, draw conclusions from the results, and disseminate the resulting scientific findings. In this CRADA work, we propose to test and validate the real-world application of the Latimer Control solution in an HIL environment. We evaluate the performance of different flexible solar technologies in responding to automatic generation control signals in a closed-loop fashion. In particular, a data-driven potential high limit (PHL) estimation is developed for large solar plants to accurately estimate their headroom so that they have fast and short-time regulation and control capability to participate in grid services and respond to grid signals in real time (e.g., AGC). This PHL estimation algorithm is embedded in a hardware power plant controller (PPC) and tested with an IEEE-39 bus system model developed in RTDS. To account for the varying cloud conditions and diverse inverter dispatches, we developed a 135-MW PV plant with detailed modeling of 27 individual PV modules and inverters using RTDS. The real-world communications used in such big plants, such as ModBus TCP/IP for inverter level and DNP3 for plant level, were developed to emulate the real-world applications in big PV plants. The ML-based PHL estimation method is tested under nine separate weather scenarios against the ‘reference-control’ solution, hereafter referred to as the baseline solution. The baseline method reserves a subset of inverters (reference group) to operate at their PHL at all times and dispatches only the remaining inverters (control group) at curtailed levels to fulfill the flexibility need. Despite being successfully piloted by NREL in California in 2017 and Chile in 2020, there exist two gaps in the state of the art to fully unlock the flexibility of PV plants: a. There is a trade-off between the PHL estimation accuracy and the flexibility range. b. There lacks granularity in the PHL estimation to capture the variation across inverters. The Latimer solution seeks to address these gaps by applying machine learning methods to improve PHL estimation accuracy while accounting for variability at every inverter. Performance metrics were taken from the 2023 Georgia Power CARES utility-scale RFP. The results demonstrate that the ML-based approach outperforms the traditional baseline method in PHL estimation accuracy for 7 of 9 scenarios. The average PHL error across the nine scenarios was 7.40% for the ML-based method, 2.06% less than the 9.46% PHL error average across scenarios that was exhibited by the baseline method. Additionally, the PHL error was below 5% for at least 95% of the testing interval for 3 of 9 tested intervals with the ML approach, whereas it did not achieve this metric for any of the baseline tests. Overall, simulation results indicate the superior performance of an ML-based approach compared to the conventional baseline reference-control approach, showcasing its potential to support grid stability and operational efficiency. This laboratory HIL testing using real PPC, representative power system simulation models in real-time with detailed PV plant and inverter models, and real-world communication protocols gives us confidence that this machine learning based PHL estimation algorithm works well in the hardware PPC and therefore de-risks future field commissioning. The end goal of this project is to advance grid technology to address the grid operation challenges brought by solar plant’s variability and uncertainties in power generation.

14 SOLAR ENERGY↗
DOE OSTI2024

Fermions at finite density in the path integral approach

Abstract We study relativistic fermionic systems in 3 + 1 spacetime dimensions at finite chemical potential and zero temperature, from a path-integral point of view. We show how to properly account for theiεterm that projects on the finite density ground state, and compute the path integral analytically for free fermions in homogeneous external backgrounds, using complex analysis techniques. As an application, we show that the U(1) symmetry is always linearly realized for free fermions at finite charge density, differently from scalars. We study various aspects of finite density QED in a homogeneous magnetic background. We compute the free energy density, non-perturbatively in the electromagnetic coupling and the external magnetic field, obtaining the finite density generalization of classic results of Euler-Heisenberg and Schwinger. We also obtain analytically the magnetic susceptibility of a relativistic Fermi gas at finite density, reproducing the de Haas-van Alphen effect. Finally, we consider a (generalized) Gross-Neveu model forNinteracting fermions at finite density. We compute its non-perturbative effective potential in the large-Nlimit, and discuss the fate of the U(1) vector and$$ {\mathbb{Z}}_2^A $$ ℤ 2 A axial symmetries.

Physics↗
DOE OSTI2024

Modeling direct air carbon capture and storage in a 1.5 °C climate future using historical analogs

Limiting the rise in global temperature to 1.5 °C will rely, in part, on technologies to remove CO 2 from the atmosphere. However, many carbon dioxide removal (CDR) technologies are in the early stages of development, and there is limited data to inform predictions of their future adoption. Here, we present an approach to model adoption of early-stage technologies such as CDR and apply it to direct air carbon capture and storage (DACCS). Our approach combines empirical data on historical technology analogs and early adoption indicators to model a range of feasible growth pathways. We use these pathways as inputs to an integrated assessment model (the Global Change Analysis Model, GCAM) and evaluate their effects under an emissions policy to limit end-of-century temperature change to 1.5 °C. Adoption varies widely across analogs, which share different strategic similarities with DACCS. If DACCS growth mirrors high-growth analogs (e.g., solar photovoltaics), it can reach up to 4.9 GtCO 2 removal by midcentury, compared to as low as 0.2 GtCO 2 for low-growth analogs (e.g., natural gas pipelines). For these slower growing analogs, unabated fossil fuel generation in 2050 is reduced by 44% compared to high-growth analogs, with implications for energy investments and stranded assets. Residual emissions at the end of the century are also substantially lower (by up to 43% and 34% in transportation and industry) under lower DACCS scenarios. The large variation in growth rates observed for different analogs can also point to policy takeaways for enabling DACCS.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2024

Compiled properties of nucleonic matter and nuclear and neutron star models from nonrelativistic and relativistic interactions

Here, this paper compiles the model parameters and zero-temperature properties of an extensive collection of published theoretical nuclear interactions, including 255 nonrelativistic (Skyrme-like) forces, 270 relativistic mean field (RMF) and point-coupling (RMF-PC) forces, and 13 Gogny-like forces. This forms the most exhaustive tabulation of model parameters to date. The properties of uniform symmetric matter and pure neutron matter at the saturation density are determined. Symmetry properties found from the second-order term of a Taylor expansion in neutron excess are compared with the energy difference of pure neutron and symmetric nuclear matter at the saturation density. Selected liquid-droplet model parameters, including the surface tension and surface symmetry energy, are determined for semi-infinite surfaces. Theoretical liquid droplet model neutron skin thicknesses of the neutron-rich closed-shell nuclei 48 Ca and 208 Pb are compared with published theoretical Hartree-Fock and experimental results. A similar comparison is made between theoretical liquid-droplet and experimental values of dipole polarizabilities. In addition, radii, binding energies, moments of inertia and tidal deformabilities of 1.2⁢M ⊙ , 1.4⁢M ⊙ and 1.6⁢M ⊙ neutron stars are computed. An extensive correlation analysis of bulk matter, nuclear structure, and low-mass neutron star properties is performed and compared with nuclear experiments and astrophysical observations.

nuclear astrophysics↗
DOE OSTI2023

Semi-Supervised, Non-Intrusive Disaggregation of Nodal Load Profiles With Significant Behind-the-Meter Solar Generation

It is of imperative interests for regional transmission organizations (RTOs) to effectively extract actual load profiles at transmission nodes with significant behind-the-meter solar generation, which remains a gap in the existing technology paradigm. This paper proposes an explicit yet efficient linear estimator to disaggregate actual load profiles at transmission buses with significant behind-the-meter (BTM) solar generations. The proposed estimator is based on disaggregating (i.e., extracting) at locations close to transmission buses under consideration. Further, to overcome the lack of “ground truth” and validate the performance of the proposed algorithms, we first propose semi-supervised mechanisms with parameter tuning as well as unsupervised clustering and leverage the unique characteristics of zero-crossing points in BTM solar peaking behaviors, which we refer to as “Zone-to-Node (Z2N)” methods. Next, we further propose a bi-level Node-to-Node (N2N) framework that improves the overall disaggregation performances compared to Z2N. Numerical results are presented using real-world data at PJM Interconnection.

14 SOLAR ENERGY↗
DOE OSTI2021

Combined First-Principles Calculations of Electron–Electron and Electron–Phonon Self-Energies in Condensed Systems

In this work, we present a method to efficiently combine the computation of electron–electron and electron–phonon self-energies, which enables the evaluation of electron–phonon coupling at the G 0 W 0 level of theory for systems with hundreds of atoms. In addition, our approach, which is a generalization of a method recently proposed for molecules [J. Chem. Theory Comput.2018, 14, 6269–6275], enables the inclusion of nonadiabatic and temperature effects at no additional computational cost. We present results for diamond and defects in diamond and discuss the importance of numerically accurate G 0 W 0 band structures to obtain robust predictions of zero point renormalization (ZPR) of band gaps, and of the inclusion of nonadiabatic effects to accurately compute the ZPR of defect states in the band gap.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Filtering, averaging, and scale dependency in homogeneous variable density turbulence

We investigate relationships between statistics obtained from filtering and from ensemble or Reynolds-averaging turbulence flow fields as a function of length scale. Generalized central moments in the filtering approach are expressed as inner products of generalized fluctuating quantities, q ' ( ξ , x ) = q ( ξ ) - q ¯ ( x ) , representing fluctuations of a field q ( ξ ) , at any point ξ, with respect to its filtered value at x. For positive-definite filter kernels, these expressions provide a scale-resolving framework, with statistics and realizability conditions at any length scale. In the small-scale limit, scale-resolving statistics become zero. In the large-scale limit, scale-resolving statistics and realizability conditions are the same as in the Reynolds-averaged description. Using direct numerical simulations (DNS) of homogeneous variable density turbulence, we diagnose Reynolds stresses, T i j , resolved kinetic energy, kr, turbulent mass-flux velocity, a i , and density-specific volume covariance, b, defined in the scale-resolving framework. These variables, and terms in their governing equations, vary smoothly between zero and their Reynolds-averaged definitions at the small and large scale limits, respectively. At intermediate scales, the governing equations exhibit interactions between terms that are not active in the Reynolds-averaged limit. For example, in the Reynolds-averaged limit, b follows a decaying process driven by a destruction term; at intermediate length scales, it is a balance between production, redistribution, destruction, and transport, where b grows as the density spectrum develops and then decays when mixing becomes strong enough. This work supports the notion of a generalized, length-scale adaptive model that converges to DNS at high resolutions and to Reynolds-averaged statistics at coarse resolutions.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2021

Energy Efficient Waste Heat Coupled Forward Osmosis for Effluent Water Management at Coal-Fired Power Plants

This project sought to evaluate the technical and economic viability of the Aquapod©, a transformational low energy (<200 kJ/kg water) waste heat coupled forward osmosis (FO) technology, to manage effluents, meet cooling water demands, and achieve water conservation in a coal-fired power plant environment. The Aquapod© process is innovative because it is heat-driven, avoids the evaporation of water, and uses no toxic chemicals such as ammonia or amines. The Aquapod© process accomplishes this using an aqueous two-phase system (ATPS) coupled FO process. The evaluation revealed that the Aquapod© process offers a pathway to exploit waste heat resources within a power plant to achieve flue-gas desulfurization (FGD) wastewater volume reduction and water recovery with minimal pretreatment. Water recovery of 80% from FGD wastewater was achieved with minimal pretreatment, exceeding the project target of 50%. The estimated electrical energy of 2.16 kWh/m 3 of water produced for the Aquapod© process met the project target of < 3.6 kWh/m 3 . The heat required for the process operation was approximately 186 kJ/kg of produced water, which was slightly lower than the project target of 200 kJ/kg. The estimated treatment cost for installing and operating the Aquapod© process in conjunction with a spray dryer evaporator to achieve zero-liquid discharge (ZLD) of the 100 gpm FGD wastewater was $13.24/m 3 over a 30-year lifetime. As a point of reference, this study’s host power plant currently incurs a cost of $3.70/m 3 – $8.87/m3 on a discharged volume basis to discharge its wastewater into the publicly owned treatment works (POTW) after physical-chemical treatment. Therefore, the Illinois power plant would incur an incremental cost increase of $4.37/m 3 – 9.54/m 3 to achieve ZLD using the Aquapod© and spray dryer combination. Ample opportunities exist to further lower the ZLD capital and operating costs in the next design iteration to attain pipe parity at the higher end of the site treatment costs.

01 COAL, LIGNITE, AND PEAT↗
DOE OSTI2022

Superconductivity out of a non-Fermi liquid: Free energy analysis

In this paper, we present an in-depth analysis of the condensation energy E c for a superconductor in a situation when superconductivity emerges out of a non-Fermi liquid due to pairing mediated by a massless boson. This is the case for electronic-mediated pairing near a quantum-critical point in metal, for pairing in SYK-type models, and for phonon-mediated pairing in the properly defined limit, when the dressed Debye frequency vanishes. We consider a subset of these quantum-critical models, in which the pairing in a channel with a proper spatial symmetry is described by an effective 0 + 1 dimensional model with the effective dynamical interaction V(Ω m ) = g¯ γ /|Ω m | γ , where γ is model-specific (the γ model). In previous papers, we argued that the pairing in the γ model is qualitatively different from that in a Fermi liquid, and the gap equation at T = 0 has an infinite number of topologically distinct solutions, Δ n (ω m ), where an integer n, running between 0 and infinity, is the number of zeros of Δ n (ω m ) on the positive Matsubara axis. This gives rise to the set of extrema of E c at E c,n , of which E c,0 is the global minimum. Here, the spectrum E c,n is discrete for a generic γ < 2 but becomes continuous at γ = 2–0. Here, we discuss in more detail the profile of the condensation energy near each E c,n and the transformation from a discrete to a continuous spectrum at γ → 2. We also discuss the free energy and the specific heat of the γ model in the normal state.

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2026

Electrolyte Organization Leads to Potential-Dependence in Thermochemical Catalysis of Nonpolar Reactions

Electrochemical polarization is now known to play a key role in thermochemical catalysis at solid–liquid interfaces. However, existing frameworks cannot account for why even nonpolar, nonfaradaic reactions are sensitive to interfacial polarization. In order to uncover the molecular basis of this phenomenon, we herein study the potential-dependent reaction kinetics of ethylene and trans-2-butene hydrogenation at Pt–liquid interfaces. Measurements were performed in aqueous and ortho-difluorobenzene (o-DFB) solutions, spontaneously polarizing the Pt–liquid interfaces by, respectively, varying the pH or dissolving distinct metallocene redox buffers into solution. Here, we find that at comparable mechanistic regimes, the rates of both ethylene and trans-2-butene hydrogenation are maximized near the same electrochemical potential, E. Moreover, the potential-dependence, defined as $\frac{∂ln 𝑟}{∂𝐸}$, of trans-2-butene hydrogenation is approximately 2.2× greater than that of ethylene hydrogenation across the full potential range studied. These observations are all consistent with a model in which polarization of the Pt surface away from the local potential of zero free charge (E PZFC ) induces electrostatic organization of the polar solvent and charged ions near the interface, which impedes olefin adsorption and surface reaction because these surface reactions induce electrolyte displacement. Accordingly, interfacial polarization alters the free energy landscape and thus the rate of nonpolar heterogeneous catalysis by controlling the degree of electrostatic organization of polar and charged spectators at the interface, which do not in general need to be specifically chemisorbed onto the surface but could simply be close enough to the surface to be perturbed by the olefin adsorption. These results point toward electrochemical design handles, namely, the electrolyte, catalyst potential, and local E PZFC of the catalyst, with which to tune interfacial catalysis of thermochemical organic transformations.

adsorption↗
DOE OSTI2022

A first-principles investigation of point defect structure and energetics in ThO 2

The structure and energetics of charged point defects in thorium dioxide (ThO 2 ) have been investigated using the density functional theory (DFT) and phonon simulations. DFT simulations were performed under both zero-pressure and constant volume conditions. Termed as the free volume change of the point defects, the change in volume of the supercell has been computed in the zero-pressure case. Supercell expansion was observed with the increase of the (nominal) charge state of anion (O) interstitials and cation (Th) vacancies from neutral to its maximum. On the contrary, contraction of the supercell has been observed with anion vacancies and cation interstitials as the defect charge increases. The supercell volume change with respect to the charge state has been correlated with the resulting defect energetics. It has been observed that, as the defect charge increased, the internal energy and entropy of defect formation of the cation vacancies and anion interstitials were found to increase, while that of the cation interstitials and anion vacancies decreased. Here, the temperature dependence of internal energy and entropy has also been examined. It was found that, as the temperature increases, the internal energies of the formation of cation vacancies and anion interstitials decrease, while those of the cation interstitials and anion vacancies increase. An opposite observation is seen for the entropies of formation defects when above room temperatures.

42 ENGINEERING↗
DOE OSTI2023

Theoretical Kinetics Predictions for Reactions on the NH 2 O Potential Energy Surface

Recent modeling studies of ammonia oxidation, which are motivated by the prospective role of ammonia as a zero-carbon fuel, have indicated significant discrepancies among the existing literature mechanisms. In this study, high-level theoretical kinetics predictions have been obtained for reactions on the NH 2 O potential energy surface, including the NH 2 + O, HNO + H, and NH + OH reactions. These reactions have previously been highlighted as important reactions in NH 3 oxidation with high sensitivity and high uncertainty. The potential energy surface is explored with coupled cluster calculations, including large basis sets and high-level corrections to yield high-accuracy (~0.2 kcal/mol 2σ uncertainty) estimates of the stationary point energies. Variational transition state theory is used to predict the microcanonical rate constants, which are then incorporated in master equation treatments of the temperature- and pressure-dependent kinetics. For radical–radical channels, the microcanonical rates are obtained from variable reaction coordinate transition state theory implementing directly evaluated multireference electronic energies. The analysis yields predictions for the total rate constants as well as the branching ratios. We find that the NO + H 2 channel contributes 10% of the total NH 2 + O flux at combustion temperatures, although this channel is not included in modern NH 3 oxidation mechanisms. Modeling is used to illustrate the ramifications of these rate predictions on the kinetics of NH 3 oxidation and NO x formation. Finally, the present results for NH 2 + O are important for predicting the chain branching and formation of NO in the oxidation of NH 3 and thermal DeNO x .

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

Julia as a unifying end-to-end workflow language on the Frontier exascale system

We evaluate Julia as a single language and ecosystem paradigm powered by LLVM to develop workflow components for high-performance computing. We run a Gray-Scott, 2-variable diffusion-reaction application using a memory-bound, 7-point stencil kernel on Frontier, the US Department of Energy’s first exascale supercomputer. We evaluate the performance, scaling, and trade-offs of (i) the computational kernel on AMD’s MI250x GPUs, (ii) weak scaling up to 4,096 MPI processes/GPUs or 512 nodes, (iii) parallel I/O writes using the ADIOS2 library bindings, and (iv) Jupyter Notebooks for interactive analysis. Results suggest that although Julia generates a reasonable LLVM-IR, a nearly 50% performance difference exists vs. native AMD HIP stencil codes when running on the GPUs. As expected, we observed near-zero overhead when using MPI and parallel I/O bindings for system-wide installed implementations. Consequently, Julia emerges as a compelling high-performance and high-productivity workflow composition language, as measured on the fastest supercomputer in the world.

Godoy, William↗