Search NASA⌕ Search

SEARCH · Search NASA

Results for “GPU power”

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 199 records · Page 11
All sourcesNASA NTRSDOE OSTI
DOE OSTI2020

TeraChem Cloud: A High-Performance Computing Service for Scalable Distributed GPU-Accelerated Electronic Structure Calculations

The encapsulation and commoditization of electronic structure arise naturally as interoperability, and the use of nontraditional compute resources (e.g., new hardware accelerators, cloud computing) remains important for the computational chemistry community. Here, we present TERACHEM CLOUD, a high-performance computing service (HPCS) that offers on-demand electronic structure calculations on both traditional HPC clusters and cloud-based hardware. The framework is designed using off-the-shelf web technologies and containerization to be extremely scalable and portable. Within the HPCS model, users can quickly develop new methods and algorithms in an interactive environment on their laptop while allowing TERACHEM CLOUD to distribute ab initio calculations across all available resources. This approach greatly increases the accessibility of hardware accelerators such as graphics processing units (GPUs) and flexibility for the development of new methods as additional electronic structure packages are integrated into the framework as alternative backends. Cost-performance analysis indicates that traditional nodes are the most cost-effective long-term solution, but commercial cloud providers offer cutting-edge hardware with competitive rates for short-term large-scale calculations. We demonstrate the power of the TERACHEM CLOUD framework by carrying out several showcase calculations, including the generation of 300,000 density functional theory energy and gradient evaluations on medium-sized organic molecules and reproducing 300 fs of nonadiabatic dynamics on the B800-B850 antenna complex in LH2, with the latter demonstration using over 50 Tesla V100 GPUs in a commercial cloud environment in 8 h for approximately $1250.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

Development of Carbon Molecular Sieves Hollow Fiber Membranes based on Polybenzimidazole Doped with Polyprotic Acids with Superior H 2 /CO 2 Separation Properties (Final Report)

The goal of this project was to develop a highly efficient membrane-based process to capture CO 2 from coal-derived syngas with 95% CO 2 purity, achieving the cost of electricity (COE) 30% below the baseline capture approaches (i.e., Selexol process) when coupled with the advancement in other areas of the power generation facility. Our core approach is based on high-permeance hollow fiber membranes (HFMs) with superior H 2 /CO 2 separation properties at the syngas process conditions, which can then be further utilized to design membrane reactors for process intensification of H 2 production and purification and CO 2 capture. Three organizations with complementary skills collaborated to achieve the goal, including the University at Buffalo (UB), Los Alamos National Laboratory (LANL), and Trimeric Corporation (Trimeric). We formulated logical steps to bring the membrane technology from Technology Readiness Level (TRL) 3 (Experimental proof of concept) to TRL 4 (Laboratory scale validation in relevant environment). During the budget period (BP) 1, we screened various polymeric materials and identified polybenzimidazole doped with inorganic polyprotic acids as the desirable platform. The acid doping increases the H 2 /CO 2 selectivity, and the sequential carbonization increases H 2 permeability while retaining the high selectivity. By manipulating the acid type and doping level and the carbonization temperature, we demonstrated advanced carbon molecular sieving (CMS) materials with H 2 permeability of above 200 Barrer (1 Barrer = 10 -10 cm 3 (STP) cm cm -2 s -2 cmHg -1 ) and H 2 /CO 2 selectivity of above 40 at 200-300°C with simulated syngas containing CO and water vapor. For example, the PBI-(H 3 PO 4 ) 0.11 carbonized at 700 °C exhibits H 2 permeability of 200 Barrer and H 2 /CO 2 selectivity of 60 at 200 °C, and H 2 permeability of 240 Barrer and H 2 /CO 2 selectivity of 54 at 225 °C, which meets the targeted properties and surpasses Robeson’s upper bound. During the BP2, we focused on the conversion of the advanced CMS materials to stable HFMs. Membranes with H 2 permeance of 1,090 GPU (1 GPU = 10 -6 cm 3 (STP) cm -2 s -2 cmHg -1 ) and H 2 /CO 2 selectivity of 57 at 300 °C were successfully fabricated. The effects of temperature, gas compositions, pressure, and time on the separation properties were systematically investigated. Pencil modules were continuously evaluated for 219 h (dry pure gas) and 669 h (dry simulated syngas) and showed initial decline in permeance and increased H 2 /CO 2 selectivity over time, ultimately achieving a steady state stable value, indicating that the ageing phenomena in the nanoporous structures of the membranes during the test. Membrane reactors were fabricated based on the CMS membranes and evaluated for water-gas shift (WGS) reaction. The use of membranes slightly improves the conversion of the CO. However, parametric tests of the membrane reactors at various temperatures and flow rates need to be conducted, as well as the membranes with improved separation performance. We performed a sensitivity analysis on the impact of H 2 /CO 2 selectivity on the COE based on a hybrid process of a membrane unit and cryogenic unit developed by Membrane Technology and Research, Inc. (MTR). Three H 2 /CO 2 selectivity (40, 60, and 15) cases were developed and compared with the baseline capture process (Case B5B) provided by the DOE report. Increasing the membrane H 2 /CO 2 selectivity reduces COE, but the rate of the decrease of COE also diminishes. The COE values for H 2 /CO 2 selectivities of 40 and 60 were nearly the same. As the H 2 /CO 2 selectivity increases, the inert recycling decreases, leading to smaller equipment, less auxiliary power requirements, and less heating, cooling, and refrigeration duty. The refrigeration system used to liquefy the CO 2 is the most expensive piece of equipment and consumes the most electricity within the CO 2 capture process. Increasing the CO 2 concentration in the recycle stream would improve the economics of the process by reducing the refrigeration duty requirement of the unit and also allow for higher liquefaction temperatures. The high H 2 -selective membrane developed by our team may be applicable in other separation processes where lower pressure H 2 retains value. Typically, hydrogen retains its pressure when it is separated from syngas components. Residual components may be used as low-quality fuel and then vented to the atmosphere. Applications might include control of H 2 /CO ratios or mitigation of the water gas shift reaction by CO 2 recycling to the feed of a gasifier or steam methane reformer. To summarize, we have developed industrial HFMs with the best H 2 /CO 2 separation performance reported in the literature. The membranes demonstrate stability with simulated syngas and show great potential for membrane reactors for WGS reactions, lowering the cost of blue H 2 production.

20 FOSSIL-FUELED POWER PLANTS↗
DOE OSTI2020

OpenABLext: An automatic code generation framework for agent-based simulations on CPU-GPU-FPGA heterogeneous platforms

The execution of agent-based simulations (ABSs) on hardware accelerator devices such as graphics processing units (GPUs) has been shown to offer great performance potentials. However, in heterogeneous hardware environments, it can become increasingly difficult to find viable partitions of the simulation and provide implementations for different hardware devices. To automate this process, we present OpenABLext, an extension to OpenABL, a model specification language for ABSs. By providing a device-aware OpenCL backend, OpenABLext enables the co-execution of ABS on heterogeneous hardware platforms consisting of central processing units, GPUs, and field programmable gate arrays (FPGAs).We present a novel online dispatching method that efficiently profiles partitions of the simulation during run-time to optimize the hardware assignment while using the profiling results to advance the simulation itself. In addition, OpenABLext features automated conflict resolution based on user-specified rules, supports graph-based simulation spaces, and utilizes an efficient neighbor search algorithm. We show the improved performance of OpenABLext and demonstrate the potential of FPGAs in the context of ABS. We illustrate how co-execution can be used to further lower execution times. OpenABLext can be seen as an enabler to tap the computing power of heterogeneous hardware platforms for ABS.

97 MATHEMATICS AND COMPUTING↗
NASA NTRSDate not supplied

Presentation Summary: State of the AGN: Progress Toward Understanding Black Hole Accretion Processes

Accretion of plasma onto black holes power some of the most powerfulsystems in the cosmos. Supermassive black holes at the centers ofgalaxies represent the high-mass limit of these objects and so accountfor the most luminous accretors. As a result, their influence spansvast spatial and temporal scales of cosmic phenomena: intraclusterheating, intergalactic media, galactic feedback and star formation,kiloparsec-scale jets/outflows, variability over time scales of minutesto centuries, and luminous multi-wavelength electromagnetic emissionextending all the way down to its event horizon. Their intrigue isheightened by the fact that they lie at the intersection of variousphysical laws---e.g., general relativistic gravity,magnetohydrodynamics, radiation, high-energy particle physics,thermodynamics, and photo-ionization---which all must be reconciled toarrive at a fundamental understanding and probe for new physics, liketests of general relativity. These physics ingredients must beincorporated into simulations performed somehow on dynamical scalesranging from that of the event horizon to parsec-scales. Fortunately,new computational and theoretical techniques---such as GPU computing,radiation transport, and and novel gridding techniques---are enablingprogress to larger scales, more degrees of freedom, and even to binarysystems. Some of the topics we will survey include recent progress onsimulating the relationship between the disk-jet interaction, howtilted black holes behave, radiation-dominated flow, and how binaryAGN affect the standard picture of black hole accretion. Along theway, we will highlight how new technologies have enabled thesescientific rewards. Future directions and open questions will beprovided to inspire discussion and interaction during the session.

Black Holes↗
DOE OSTI2026

Center for Integrated Simulation of Fusion Relevant RF Actuators

This project was part of the “Center for Integrated Simulation of Fusion Relevant RF Actuators” SciDAC-4 project, led by Dr. Paul Bonoli (MIT). Rather than use an acronym (CISFRRFA), the project will be referred to in this document as the “RF-SciDAC4”. The larger SciDAC-4 project goals were to: 1. Develop an integrated simulation of the antenna + sheath + scrape-off-layer + core plasma system which fully utilizes leadership class computing resources to move towards a quantitative predictive capability for the response to RF power. 2. Work closely with the SciDAC-4 Whole Device Modeling (WDM) community to make both our new code development efforts, as well as the established hierarchy of RF tools, available within their environment, and to utilize WDM technologies to implement the couplings below. 3. Validate this predictive capability on appropriately diagnosed experiments including dedicated RF test stands, linear devices, and existing tokamaks. 4. Use these tools to inform design of robust, impurity-mitigating RF heating and current drive sources for future fusion devices.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2023

Methane Emissions Produced by Pneumatic Devices and Produced Water Tanks on Natural Gas Wellsites

In recent years, more scrutiny has been placed on the release of greenhouse gasses (GHGs) by the oil and gas industry in North America as climate concerns increase. In the coming years, stricter regulations proposed by the Environmental Protection Agency (EPA) concerning the release of GHGs will further direct attention to GHG emissions quantification, modeling, and overall reduction. Active natural gas wellsites employ production equipment that routinely emits methane during normal operation. Production equipment often includes, but is not limited to, gas production units (GPUs), pneumatic controllers (PCs) and actuators powered by produced natural gas, and produced water storage tanks. This work details direct methane emission measurements taken at active natural gas wellsites originating from intermittent pneumatic controllers and produced water storage tanks. Efforts to model methane flows originating from these sources are also detailed, and those results are compared to direct measurement. The direct methane emissions measurement campaign collected flow data from two active dry wellsites in the Marcellus shale region. These measurement campaigns captured, in total, several weeks of continuous data from these sites. Data included multiple pneumatic device vent flowrate channels, a single channel monitoring total emissions vented from the produced water storage tanks, and corresponding site weather data. Periodic composition measurements were taken from flows vented from produced water tanks. Computational modeling efforts attempted to mimic a given site’s methane emissions behavior given limited input parameters. Direct measurement data and simulation data yielded, through reduction and analysis, apparent methane emissions factors, statistics on pneumatic device actuation events, important relationships between GPU parameters and pneumatic device actuation events, produced water storage tank vent flows and flow compositions, and parameters most contributing to the magnitude and frequency of these fugitive methane emissions. Measurements, considering data from a 72hr period, produced PC emissions factors from 0.008 to 0.03 SCFH per device with an average device emitting 0.026 SCFH at the first site. The second site produced PC emissions factors spanning from 0.05 to 25.47 SCFH with an average device emitting 1.93 SCFH not including the 25.47 SCFH device, which was assumed to be malfunctioning. Two unique water tank emission measurements were made, 618 g CH4 per hour and 32.6 g CH4 per hour. On a water basis, the water tank emissions factors were 0.66 and 0.042 kg CH4 per bbl of water, respectively.

03 NATURAL GAS↗
NASA NTRSDate not supplied

State of the AGN: Progress Toward Understanding Black Hole Accretion Processes

Accretion of plasma onto black holes power some of the most powerful systems in the cosmos. Supermassive black holes at the centers of galaxies represent the high-mass limit of these objects and so account for the most luminous accretors. As a result, their influence spans vast spatial and temporal scales of cosmic phenomena: intracluster heating, intergalactic media, galactic feedback and star formation, kiloparsec-scale jets/outflows, variability over time scales of minutes to centuries, and luminous multi-wavelength electromagnetic emission extending all the way down to its event horizon. Their intrigue is heightened by the fact that they lie at the intersection of various physical laws---e.g., general relativistic gravity, magnetohydrodynamics, radiation, high-energy particle physics, thermodynamics, and photo-ionization---which all must be reconciled to arrive at a fundamental understanding and probe for new physics, like tests of general relativity. These physics ingredients must be incorporated into simulations performed somehow on dynamical scales ranging from that of the event horizon to parsec-scales. Fortunately, new computational and theoretical techniques---such as GPU computing, radiation transport, and and novel gridding techniques---are enabling progress to larger scales, more degrees of freedom, and even to binary systems. Some of the topics we will survey include recent progress on simulating the relationship between the disk-jet interaction, how tilted black holes behave, radiation-dominated flow, and how binary AGN affect the standard picture of black hole accretion. Along the way, we will highlight how new technologies have enabled these scientific rewards. Future directions and open questions will be provided to inspire discussion and interaction during the session.

LISA↗
DOE OSTI2026

pop-cosmos : redshifts and physical properties of KiDS-1000 galaxies

ABSTRACT Principled Bayesian inference of galaxy properties has not previously been performed for wide-area weak-lensing surveys with millions of sources. We address this gap by applying the pop-cosmos generative model to perform spectral energy distribution (SED) fitting for 4 million KiDS (Kilo-Degree Survey)-1000 galaxies. Calibrated on deep COSMOS2020 photometric data, pop-cosmos specifies a physically motivated prior over the galaxy population up to $z \simeq 6$ in stellar population synthesis (SPS) parameter space. Using the Speculator SPS emulator with GPU (graphics processing unit)-accelerated Markov Chain Monte Carlo sampling, we perform full posterior inference at 8.2 GPU seconds per galaxy, obtaining joint constraints on galaxy redshifts and physical properties. We validate photometric redshifts against $\sim \!185\,\!000$ KiDS galaxies cross-matched to Dark Energy Spectroscopic Instrument Data Release 1 spectroscopic samples, achieving low bias ($2\times 10^{-3}$), scatter ($\sigma _{\mathrm{MAD}}=0.03$), and outlier fraction (3.2 per cent) for the Bright Galaxy Survey, with comparable performance (bias $3\times 10^{-2}$, $\sigma _{\mathrm{MAD}}=0.05$, 1.0 per cent outliers) for luminous red galaxies (LRGs). Within the LRG sample, we identify massive, dusty, star-forming contaminants at $z \simeq 0.4$ satisfying standard colour selections for quenched populations. We infer trends in stellar mass, star formation, metallicity, and dust across five tomographic redshift bins consistent with established scaling relations. Using specific star formation rate constraints, we identify $\sim$7 per cent of KiDS-1000 galaxies as quenched, versus 37 per cent implied by conservative colour cuts. This enables the construction of weak-lensing samples defined by physical properties while mitigating intrinsic alignment systematics and preserving statistical power. Our analysis validates pop-cosmos out of sample, establishing it as a scalable approach for galaxy evolution and cosmological analyses with photometric surveys.

Halder, Anik [Institute of Astronomy and Kavli Ins↗
DOE OSTI2022

BUTTER-E - Energy Consumption Data for the BUTTER Empirical Deep Learning Dataset

The BUTTER-E - Energy Consumption Data for the BUTTER Empirical Deep Learning Dataset adds node-level energy consumption data from watt-meters to the primary sweep of the BUTTER - Empirical Deep Learning Dataset. This dataset contains energy consumption and performance data from 63,527 individual experimental runs spanning 30,582 distinct configurations: 13 datasets, 20 sizes (number of trainable parameters), 8 network "shapes", and 14 depths on both CPU and GPU hardware collected using node-level watt-meters. This dataset reveals the complex relationship between dataset size, network structure, and energy use, and highlights the impact of cache effects. BUTTER-E is intended to be joined with the BUTTER dataset (see "BUTTER - Empirical Deep Learning Dataset on OEDI" resource below) which characterizes the performance of 483k distinct fully connected neural networks but does not include energy measurements.

Array↗
DOE OSTI2020

Heterogeneous Reconstruction of Tracks and Primary Vertices With the CMS Pixel Tracker

The High-Luminosity upgrade of the Large Hadron Collider (LHC) will see the accelerator reach an instantaneous luminosity of 7 × 10 34 cm −2 s −1 with an average pileup of 200 proton-proton collisions. These conditions will pose an unprecedented challenge to the online and offline reconstruction software developed by the experiments. The computational complexity will exceed by far the expected increase in processing power for conventional CPUs, demanding an alternative approach. Industry and High-Performance Computing (HPC) centers are successfully using heterogeneous computing platforms to achieve higher throughput and better energy efficiency by matching each job to the most appropriate architecture. In this paper we will describe the results of a heterogeneous implementation of pixel tracks and vertices reconstruction chain on Graphics Processing Units (GPUs). The framework has been designed and developed to be integrated in the CMS reconstruction software, CMSSW. The speed up achieved by leveraging GPUs allows for more complex algorithms to be executed, obtaining better physics output and a higher throughput.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2024

High-performance finite elements with MFEM

The MFEM (Modular Finite Element Methods) library is a high-performance C++ library for finite element discretizations. MFEM supports numerous types of finite element methods and is the discretization engine powering many computational physics and engineering applications across a number of domains. Furthermore, this paper describes some of the recent research and development in MFEM, focusing on performance portability across leadership-class supercomputing facilities, including exascale supercomputers, as well as new capabilities and functionality, enabling a wider range of applications. Much of this work was undertaken as part of the Department of Energy’s Exascale Computing Project (ECP) in collaboration with the Center for Efficient Exascale Discretizations (CEED).

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Dissecting Tensor Cores via Microbenchmarks: Latency, Throughput and Numeric Behaviors

Tensor Cores have been an important unit to accelerate Fused Matrix Multiplication Accumulation (MMA) in all NVIDIA GPUs since Volta Architecture. To program Tensor Cores, users have to use either legacy wmma APIs or current mma APIs. Legacy wmma APIs are more easy-to-use but can only exploit limited features and power of Tensor Cores. Specifically, wmma APIs support fewer operand shapes and can not leverage the new sparse matrix multiplication feature of the newest Ampere Tensor Cores. However, the performance of current programming interface has not been well explored. Furthermore, the computation numeric behaviors of low-precision floating points (TF32, BF16, and FP16) supported by the newest Ampere Tensor Cores are also mysterious. In this paper, we explore the throughput and latency of current programming APIs. Further, we intuitively study the numeric behaviors of Tensor Cores MMA and profile the intermediate operations including multiplication, addition of inner product, and accumulation. All codes used in this work can be found in https://github.com/sunlex0717/DissectingTensorCores.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Application of performance portability solutions for GPUs and many-core CPUs to track reconstruction kernels

Next generation High-Energy Physics (HEP) experiments are presented with significant computational challenges, both in terms of data volume and processing power. Using compute accelerators, such as GPUs, is one of the promising ways to provide the necessary computational power to meet the challenge. The current programming models for compute accelerators often involve using architecture-specific programming languages promoted by the hardware vendors and hence limit the set of platforms that the code can run on. Developing software with platform restrictions is especially unfeasible for HEP communities as it takes significant effort to convert typical HEP algorithms into ones that are efficient for compute accelerators. Multiple performance portability solutions have recently emerged and provide an alternative path for using compute accelerators, which allow the code to be executed on hardware from different vendors. We apply several portability solutions, such as Kokkos, SYCL, C++17 std::execution::par and Alpaka, on two mini-apps extracted from the mkFit project: p2z and p2r. These apps include basic kernels for a Kalman filter track fit, such as propagation and update of track parameters, for detectors at a fixed z or fixed r position, respectively. The two mini-apps explore different memory layout formats. We report on the development experience with different portability solutions, as well as their performance on GPUs and many-core CPUs, measured as the throughput of the kernels from different GPU and CPU vendors such as NVIDIA, AMD and Intel.

Kwok, Ka Martin↗
DOE OSTI2024

Application of performance portability solutions for GPUs and many-core CPUs to track reconstruction kernels

Next generation High-Energy Physics (HEP) experiments are presented with significant computational challenges, both in terms of data volume and processing power. Using compute accelerators, such as GPUs, is one of the promising ways to provide the necessary computational power to meet the challenge. The current programming models for compute accelerators often involve using architecture-specific programming languages promoted by the hardware vendors and hence limit the set of platforms that the code can run on. Developing software with platform restrictions is especially unfeasible for HEP communities as it takes significant effort to convert typical HEP algorithms into ones that are efficient for compute accelerators. Multiple performance portability solutions have recently emerged and provide an alternative path for using compute accelerators, which allow the code to be executed on hardware from different vendors. We apply several portability solutions, such as Kokkos, SYCL, C++17 std::execution::par, Alpaka, and OpenMP/OpenACC, on two mini-apps extracted from the mkFit project: p2z and p2r. These apps include basic kernels for a Kalman filter track fit, such as propagation and update of track parameters, for detectors at a fixed z or fixed r position, respectively. The two mini-apps explore different memory layout formats. We report on the development experience with different portability solutions, as well as their performance on GPUs and many-core CPUs, measured as the throughput of the kernels from different GPU and CPU vendors such as NVIDIA, AMD and Intel.

Kwok, Ka Hei Martin↗
DOE OSTI2021

ExaWind: Predictive Wind Energy Simulations

This presentation describes the ExaWind project and the team's progress in creating a suite of performance-portable codes designed for predictive simulations of wind farms on next-generation exascale-class supercomputers. Such simulations will require the resolution of scales spanning many orders of magnitude, from blade boundary layers to wind farm flow structures. In the U.S., the first exascale systems will be GPU accelerated, and different GPU manufacturers have been chosen for the different systems. At the heart of the ExaWind software is a hybrid-solver approach based on the codes Nalu-Wind and AMR-Wind, which are computational fluid dynamics solvers for the incompressible Navier-Stokes equations. Nalu-Wind is an unstructured-grid code used to resolve wind turbine geometry and blade boundary layers, whereas AMR-Wind is a structured-grid background solver for atmospheric turbulent flow and turbine wake propagation. The models are coupled with overset meshes and global linear systems are approximated through a loose-coupling algorithm. Results will include validation-quality high-fidelity simulations and strong/weak scaling results from the Summit supercomputer.

49 EE - Wind and Water Power Program - Wind (EE-4W↗
DOE OSTI2020

The Multiphysics on Advanced Platforms Project

In 2015, the Lawrence Livermore National Laboratory started development of next-generation multiphysics simulation capabilities for the National Nuclear Security Administration under the Advanced Technologies Development and Mitigation (ATDM) element of the Advanced Simulation and Computing program in collaboration with the Exascale Computing Project (ECP). A key driver for this effort across the NNSA tri-lab was the emergence of advanced high performance computing (HPC) architectures based on heterogeneous compute capabilities, including GPU based systems, as part of the national drive toward exascale computing platforms at multiple Department of Energy (DOE) facilities. Developing a multiphysics code capable of meeting the various simulation needs of the NNSA as defined by the current generation of integrated codes (or ICs), initially developed as part of the Accelerated Strategic Computing Initiative (ASCI) program beginning in 1996, and able to scale to the current 100 petaflop class pre-exascale systems, as well the forthcoming exaflop class computers, is a daunting challenge. To accomplish this ambitious goal, LLNL has embraced two key themes: use of high-order numerical methods and a modular approach to code development. The LLNL next generation effort is organized under the Multi-Physics on Advanced Platforms Project (MAPP). A foundational component of MAPP is the Axom computer science (CS) toolkit which provides infrastructure for the development of modular, performance portable, multi-physics application codes. MARBL is a next-generation application code built on the Axom base to address the modeling needs of the high energy density physics (HEDP) community for simulating high-explosive, magnetic or laser driven experiments such as inertial confinement fusion (ICF), pulsed-power magneto-hydrodynamics (MHD), equation of state (EOS) and material strength studies as part of the NNSA’s stockpile stewardship program (SSP).

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Unified Language Frontend for Physic-Informed AI/ML

Artificial intelligence and machine learning (AI/ML) are becoming important tools for scientific modeling and simulation as in several other fields such as image analysis and natural language processing. ML techniques can leverage the computing power available in modern systems and reduce the human effort needed to configure experiments, interpret and visualize results, draw conclusions from huge quantities of raw data, and build surrogates for physics based models. Domain scientists in fields like fluid dynamics, microelectronics and chemistry can automate many of their most difficult and repetitive tasks or improve the design times by use of the faster ML-surrogates. However, modern ML and traditional scientific highperformance computing (HPC) tend to use completely different software ecosystems. While ML frameworks like PyTorch and TensorFlow provide Python APIs, most HPC applications and libraries are written in C++. Direct interoperability between the two languages is possible but is tedious and error-prone. In this work, we show that a compiler-based approach can bridge the gap between ML frameworks and scientific software with less developer effort and better efficiency. We use the MLIR (multi-level intermediate representation) ecosystem to compile a pre-trained convolutional neural network (CNN) in PyTorch to freestanding C++ source code in the Kokkos programming model. Kokkos is a programming model widely used in HPC to write portable, shared-memory parallel code that can natively target a variety of CPU and GPU architectures. Our compiler-generated source code can be directly integrated into any Kokkosbased application with no dependencies on Python or cross-language interfaces.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2023

Development of $\mathrm{AMOEBA}$ Polarizable Force Field for Rare-Earth La 3+ Interaction with Bioinspired Ligands

Rare-earth metals (REMs) are crucial for many important industries, such as power generation and storage, in addition to cancer treatment and medical imaging. One promising new REM refinement approach involves mimicking the highly selective and efficient binding of REMs observed in relatively recently discovered proteins. However, realizing any such bioinspired approach requires an understanding of the biological recognition mechanisms. In this report we developed a new classical polarizable force field based on the AMOEBA framework for modeling a lanthanum ion (La 3+ ) interacting with water, acetate, and acetamide, which have been found to coordinate the ion in proteins. The parameters were derived by comparing to high-level ab initio quantum mechanical (QM) calculations that include relativistic effects. The AMOEBA model, with advanced atomic multipoles and electronic polarization, is successful in capturing both the QM distance-dependent La 3+ –ligand interaction energies and experimental hydration free energy. A new scheme for pairwise polarization damping (POLPAIR) was developed to describe the polarization energy in La 3+ interactions with both charged and neutral ligands. Simulations of La3+ in water showed water coordination numbers and ion–water distances consistent with previous experimental and theoretical findings. Water residence time analysis revealed both fast and slow kinetics in water exchange around the ion. This new model will allow investigation of fully solvated lanthanum ion–protein systems using GPU-accelerated dynamics simulations to gain insights on binding selectivity, which may be applied to the design of synthetic analogues.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗