Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “graphics processing units”

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 163 records · Page 9
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

COLUMBUS─An Efficient and General Program Package for Ground and Excited State Computations Including Spin–Orbit Couplings and Dynamics

The COLUMBUS program system provides the tools for performing high-level multireference (MR) computations, including the multireference configuration interaction (MRCI) method and its multireference averaged quadratic coupled cluster (MR-AQCC) extension, allowing computations on a wide range of fascinating atomic and molecular systems, including the treatment of open-shells and complicated excited state phenomena. The inclusion of spin−orbit coupling (SOC) directly within the MRCI step enables the description of systems containing heavy elements, such as lanthanides and actinides, whose properties are strongly influenced by SOC. Analytic energy gradients and nonadiabatic couplings at the correlated MRCI level provide the foundation for a variety of dynamics studies, giving insight into ultrafast photochemistry. New and ongoing method developments in COLUMBUS include the computation of spin densities, improved descriptions of ionic states, enhancements to the AQCC method, and the porting of COLUMBUS to graphical processing units (GPUs). New external interfaces enable an enhanced description of electronic resonances and molecules in strong laser fields. This work highlights these new developments while providing a detailed account of the diverse applications of COLUMBUS in recent years.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

Implementation of McMurchie–Davidson Algorithm for Gaussian AO Integrals Suited for SIMD Processors

We report an implementation of the McMurchie− Davidson evaluation scheme for 1- and 2-particle Gaussian AO integrals designed for processors with Single Instruction Multiple Data (SIMD) instruction sets. Like in our recent MD implementation for graphical processing units (GPUs) [Asadchev, A.; Valeev, E. F.. J. Chem. Phys. 2024, 160, 244109.], variable-sized batches of shellsets of integrals are evaluated at a time. By optimizing for the floating point instruction throughput rather than minimizing the number of operations, this approach achieves up to 50% of the theoretical hardware peak FP64 performance for many common SIMD-equipped platforms (AVX2, AVX512, NEON), which translates to speedups of up to 30 over the state-of-the-art one-shellset-at-a-time implementation of Obara−Saika-type schemes in Libint for a variety of primitive and contracted integrals. As with our previous work, we rely on the standard C++ programming language such as the std::simd standard library feature to be included in the 2026 ISO C++ standard without any explicit code generation to keep the code base small and portable. The implementation is part of the open source LibintX library freely available at https://github.com/ValeevGroup/libintx.

Basis sets↗
DOE OSTI2025

A GPU‐Based Ocean Dynamical Core for Routine Mesoscale‐Resolving Climate Simulations

Abstract We describe an ocean hydrostatic dynamical core implemented in Oceananigans optimized for Graphical Processing Unit (GPU) architectures. On 64 A100 GPUs, equivalent to 16 computational nodes in current state‐of‐the‐art supercomputers, our dynamical core can simulate a decade of near‐global ocean dynamics per wall‐clock day at an 8‐km horizontal resolution; a resolution adequate to resolve the ocean's mesoscale eddy field. Such efficiency, achieved with relatively modest hardware resources, suggests that climate simulations on GPUs can incorporate fully eddy‐resolving ocean models. This removes a major source of systematic bias in current IPCC coupled model projections, the parameterization of ocean eddies, and represents a major advance in climate modeling. We discuss the computational strategies, focusing on GPU‐specific optimization and numerical implementation details that enable such high performance.

Silvestri, Simone [Massachusetts Institute of Tech↗
DOE OSTI2025

ERF: Energy Research and Forecasting Model

High performance computing (HPC) architectures have undergone rapid development in recent years. As a result, established software suites face an ever increasing challenge to remain performant on and portable across modern systems. Many of the widely adopted atmospheric modeling codes cannot fully (or in some cases, at all) leverage the acceleration provided by General-Purpose Graphics Processing Units, leaving users of those codes constrained to increasingly limited HPC resources. Energy Research and Forecasting (ERF) is a regional atmospheric modeling code that leverages the latest HPC architectures, whether composed of only Central Processing Units (CPUs) or incorporating GPUs. ERF contains many of the standard discretizations and basic features needed to model general atmospheric dynamics. The modular design of ERF provides a flexible platform for exploring different physics parameterizations and numerical strategies. ERF is built on a state-of-the-art, well-supported, software framework (AMReX) that provides a performance portable interface and ensures ERF's long-term sustainability on next generation computing systems. This paper details the numerical methodology of ERF, presents results for a series of verification/validation cases, and documents ERF's performance on current HPC systems. The roughly 5× speed up of ERF (using GPUs) over Weather Research and Forecasting (CPUs only) for a 3D squall line test case highlights the significance of leveraging GPU acceleration.

17 WIND ENERGY↗
DOE OSTI2025

JAX-CanVeg: A Differentiable Land Surface Model

Land surface models consider the exchange of water, energy, and carbon along the soil-canopy-atmosphere continuum, which is challenging to model due to their complex interdependency and associated challenges in representing and parameterizing them. Differentiable modeling provides a new opportunity to capture these complex interactions by seamlessly hybridizing process-based models with deep neural networks (DNNs), benefiting both worlds, that is, the physical interpretation of process-based models and the learning power of DNNs. Here, we developed a differentiable land model, JAX-CanVeg. The new model builds on the legacy CanVeg by incorporating advanced functionalities through JAX in the graphic processing unit support, automatic differentiation, and integration with DNNs. We demonstrated JAX-CanVeg's hybrid modeling capability by applying the model at four flux tower sites with varying aridity. To this end, we developed a hybrid version of the Ball-Berry equation that emulates the water stress impact on stomatal closure to explore the capability of the hybrid model in (a) improving the simulations of latent heat fluxes (LE) and net ecosystem exchange (NEE), (b) improving the optimization trade-off when learning observations of both LE and NEE, and (c) benefiting a multi-layer canopy model setup. Our results show that the proposed hybrid model improved the simulations of LE and NEE at all sites, with an improved optimization trade-off over the process-based model. Additionally, the multi-layer canopy set benefited hybrid modeling at some sites. Anchored in differentiable modeling, our study provides a new avenue for modeling land-atmosphere interactions by leveraging the benefits of both data-driven learning and process-based modeling.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Adaptive Grid Redistribution for a 1D Model of Turbulence and Clouds

In global atmospheric models, resolving stratocumulus (Sc) in the vertical is computationally expensive. However, Sc appear only under special meteorological conditions. Therefore, there is motivation to refine the vertical grid levels adaptively. In order to facilitate the possibility of parallelization on graphical processing units, our grid adaptation method prescribes the number of vertical levels a priori. Then grid levels are relocated toward altitude ranges in need of refinement. Because the method relocates existing grid levels, rather than adding extra levels, there is a risk of creating regions with overly coarse grid spacing, that is, voids in the grid mesh. To prevent such voids from forming, a simple method is developed to impose a maximum grid spacing. To decide where to place enhanced resolution, the authors develop an empirical mesh refinement criterion. It refines grid spacing near the ground, near strong temperature gradients, and within clouds. Our grid adaptation method is implemented in a single-column model and evaluated on four test cases: decaying stratocumulus, developing shallow cumulus, a quasi-stationary stratocumulus deck, and the diurnal cycle of a dry boundary layer. In the stratocumulus cases, mesh refinement leads to improvements in both the time evolution of fields and their time averages. The other two cases show smaller differences.

Carstensen, Steffen [Univ. of Wisconsin, Milwaukee↗
DOE OSTI2024

All-optical image denoising using a diffractive visual processor

Abstract Image denoising, one of the essential inverse problems, targets to remove noise/artifacts from input images. In general, digital image denoising algorithms, executed on computers, present latency due to several iterations implemented in, e.g., graphics processing units (GPUs). While deep learning-enabled methods can operate non-iteratively, they also introduce latency and impose a significant computational burden, leading to increased power consumption. Here, we introduce an analog diffractive image denoiser to all-optically and non-iteratively clean various forms of noise and artifacts from input images – implemented at the speed of light propagation within a thin diffractive visual processor that axially spans <250 × λ, where λ is the wavelength of light. This all-optical image denoiser comprises passive transmissive layers optimized using deep learning to physically scatter the optical modes that represent various noise features, causing them to miss the output image Field-of-View (FoV) while retaining the object features of interest. Our results show that these diffractive denoisers can efficiently remove salt and pepper noise and image rendering-related spatial artifacts from input phase or intensity images while achieving an output power efficiency of ~30–40%. We experimentally demonstrated the effectiveness of this analog denoiser architecture using a 3D-printed diffractive visual processor operating at the terahertz spectrum. Owing to their speed, power-efficiency, and minimal computational overhead, all-optical diffractive denoisers can be transformative for various image display and projection systems, including, e.g., holographic displays.

36 MATERIALS SCIENCE↗
DOE OSTI2022

Scalable and accurate multi-GPU-based image reconstruction of large-scale ptychography data

Abstract While the advances in synchrotron light sources, together with the development of focusing optics and detectors, allow nanoscale ptychographic imaging of materials and biological specimens, the corresponding experiments can yield terabyte-scale volumes of data that can impose a heavy burden on the computing platform. Although graphics processing units (GPUs) provide high performance for such large-scale ptychography datasets, a single GPU is typically insufficient for analysis and reconstruction. Several works have considered leveraging multiple GPUs to accelerate the ptychographic reconstruction. However, most of these works utilize only the Message Passing Interface to handle the communications between GPUs. This approach poses inefficiency for a hardware configuration that has multiple GPUs in a single node, especially while reconstructing a single large projection, since it provides no optimizations to handle the heterogeneous GPU interconnections containing both low-speed (e.g., PCIe) and high-speed links (e.g., NVLink). In this paper, we provide an optimized intranode multi-GPU implementation that can efficiently solve large-scale ptychographic reconstruction problems. We focus on the maximum likelihood reconstruction problem using a conjugate gradient (CG) method for the solution and propose a novel hybrid parallelization model to address the performance bottlenecks in the CG solver. Accordingly, we have developed a tool, called PtyGer ( Pty chographic G PU(multipl e )-based r econstruction), implementing our hybrid parallelization model design. A comprehensive evaluation verifies that PtyGer can fully preserve the original algorithm’s accuracy while achieving outstanding intranode GPU scalability.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

HDBind: encoding of molecular structure with hyperdimensional binary representations

Traditional methods for identifying “hit” molecules from a large collection of potential drug-like candidates rely on biophysical theory to compute approximations to the Gibbs free energy of the binding interaction between the drug and its protein target. These approaches have a significant limitation in that they require exceptional computing capabilities for even relatively small collections of molecules. Increasingly large and complex state-of-the-art deep learning approaches have gained popularity with the promise to improve the productivity of drug design, notorious for its numerous failures. However, as deep learning models increase in their size and complexity, their acceleration at the hardware level becomes more challenging. Hyperdimensional Computing (HDC) has recently gained attention in the computer hardware community due to its algorithmic simplicity relative to deep learning approaches. The HDC learning paradigm, which represents data with high-dimension binary vectors, allows the use of low-precision binary vector arithmetic to create models of the data that can be learned without the need for the gradient-based optimization required in many conventional machine learning and deep learning methods. This algorithmic simplicity allows for acceleration in hardware that has been previously demonstrated in a range of application areas (computer vision, bioinformatics, mass spectrometery, remote sensing, edge devices, etc.). To the best of our knowledge, our work is the first to consider HDC for the task of fast and efficient screening of modern drug-like compound libraries. We also propose the first HDC graph-based encoding methods for molecular data, demonstrating consistent and substantial improvement over previous work. We compare our approaches to alternative approaches on the well-studied MoleculeNet dataset and the recently proposed LIT-PCBA dataset derived from high quality PubChem assays. We demonstrate our methods on multiple target hardware platforms, including Graphics Processing Units (GPUs) and Field Programmable Gate Arrays (FPGAs), showing at least an order of magnitude improvement in energy efficiency versus even our smallest neural network baseline model with a single hidden layer. Our work thus motivates further investigation into molecular representation learning to develop ultra-efficient pre-screening tools. We make our code publicly available at https://github.com/LLNL/hdbind.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

First principles reaction discovery: from the Schrodinger equation to experimental prediction for methane pyrolysis

Our recent success in exploiting graphical processing units (GPUs) to accelerate quantum chemistry computations led to the development of the ab initio nanoreactor, a computational framework for automatic reaction discovery and kinetic model construction. In this work, we apply the ab initio nanoreactor to methane pyrolysis, from automatic reaction discovery to path refinement and kinetic modeling. Elementary reactions occurring during methane pyrolysis are revealed using GPU-accelerated ab initio molecular dynamics simulations. Subsequently, these reaction paths are refined at a higher level of theory with optimized reactant, product, and transition state geometries. Reaction rate coefficients are calculated by transition state theory based on the optimized reaction paths. The discovered reactions lead to a kinetic model with 53 species and 134 reactions, which is validated against experimental data and simulations using literature kinetic models. We highlight the advantage of leveraging local brute force and Monte Carlo sensitivity analysis approaches for efficient identification of important reactions. Both sensitivity approaches can further improve the accuracy of the methane pyrolysis kinetic model. The results in this work demonstrate the power of the ab initio nanoreactor framework for computationally affordable systematic reaction discovery and accurate kinetic modeling.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Bringing heterogeneity to the CMS software framework

The advent of computing resources with co-processors, for example Graphics Processing Units (GPU) or Field-Programmable Gate Arrays (FPGA), for use cases like the CMS High-Level Trigger (HLT) or data processing at leadership-class supercomputers imposes challenges for the current data processing frameworks. These challenges include developing a model for algorithms to offload their computations on the co-processors as well as keeping the traditional CPU busy doing other work. The CMS data processing framework, CMSSW, implements multithreading using the Intel Threading Building Blocks (TBB) library, that utilizes tasks as concurrent units of work. In this paper we will discuss a generic mechanism to interact effectively with non-CPU resources that has been implemented in CMSSW. In addition, configuring such a heterogeneous system is challenging. In CMSSW an application is configured with a configuration file written in the Python language. The algorithm types are part of the configuration. The challenge therefore is to unify the CPU and co-processor settings while allowing their implementations to be separate. We will explain how we solved these challenges while minimizing the necessary changes to the CMSSW framework. We will also discuss on a concrete example how algorithms would offload work to NVIDIA GPUs using directly the CUDA API.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

Performance of Heterogeneous Algorithm Scheduling in CMSSW

The CMS experiment started to utilize Graphics Processing Units (GPU) to accelerate the online reconstruction and event selection running on its High Level Trigger (HLT) farm in the 2022 data taking period. The projections of the HLT farm to the High-Luminosity LHC foresee a significant use of compute accelerators in the LHC Run 4 and onwards in order to keep the cost, size, and power budget of the farm under control. This direction of leveraging compute accelerators has synergies with the increasing use of HPC resources in HEP computing, as HPC machines are employing more and more compute accelerators that are predominantly GPUs today. In this work we review the features developed for the CMS data processing framework, CMSSW, to support the effective utilization of both compute accelerators and many-core CPUs within a highly concurrent task-based framework. We measure the impact of various design choices for the scheduling of heterogeneous algorithms on the event processing throughput, using the Run-3 HLT application as a realistic use case.

Bocci, Andrea↗
DOE OSTI2025

Evolution of the ATLAS event data model for the HL-LHC

The upcoming high-luminosity run of the CERN Large Hadron Collider (HL-LHC) will yield an unprecedented volume of data. In order to process this data, the ATLAS collaboration is evolving its offline software to be able to use heterogeneous resources such as graphical processing units (GPUs) and field-programmable gate arrays (FPGAs). To reduce conversion overheads, the event data model (EDM) should be compatible with the requirements of these resources. While the ATLAS EDM has long allowed representing data as a structure of arrays, further evolution of the EDM can enable more efficient sharing of data between CPU and GPU resources. Some of this work will be summarized here, including extensions to allow controlling how memory for event data is allocated and the implementation of jagged vectors.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2020

Spatial core-edge coupling of the particle-in-cell gyrokinetic codes GEM and XGC

Two existing particle-in-cell gyrokinetic codes, GEM for the core region and XGC for the edge region, have been successfully coupled with a spatial coupling scheme at the interface in a toroidal geometry. Additionally, a mapping technique is developed for transferring data between GEM's structured and XGC's unstructured meshes. Two examples of coupled simulations are presented to demonstrate the coupling scheme. The optimization of GEM for graphics processing unit is also presented.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2021

Reduced scaling formulation of CASPT2 analytical gradients using the supporting subspace method

We present a reduced scaling and exact reformulation of state specific complete active space second-order perturbation (CASPT2) analytical gradients in terms of the MP2 and Fock derivatives using the supporting subspace method. This work follows naturally from the supporting subspace formulation of the CASPT2 energy in terms of the MP2 energy using dressed orbitals and Fock builds. For a given active space configuration, the terms corresponding to the MP2-gradient can be evaluated with O(N5) operations, while the rest of the calculations can be computed with O(N3) operations using Fock builds, Fock gradients, and linear algebra. When tensor-hyper-contraction is applied simultaneously, the computational cost can be further reduced to O(N4) for a fixed active space size. The new formulation enables efficient implementation of CASPT2 analytical gradients by leveraging the existing graphical processing unit (GPU)-based MP2 and Fock routines. We present benchmark results that demonstrate the accuracy and performance of the new method. Example applications of the new method in ab initio molecular dynamics simulation and constrained geometry optimization are given.

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

Toward exascale whole-device modeling of fusion devices: Porting the GENE gyrokinetic microturbulence code to GPU

GENE solves the five-dimensional gyrokinetic equations to simulate the development and evolution of plasma microturbulence in magnetic fusion devices. The plasma model used is close to first principles and computationally very expensive to solve in the relevant physical regimes. In order to use the emerging computational capabilities to gain new physics insights, several new numerical and computational developments are required. Here, we focus on the fact that it is crucial to efficiently utilize GPUs (graphics processing units) that provide the vast majority of the computational power on such systems. In this paper, we describe the various porting approaches considered and given the constraints of the GENE code and its development model, justify the decisions made, and describe the path taken in porting GENE to GPUs. We introduce a novel library called gtensor that was developed along the way to support the process. Performance results are presented for the ported code, which in a single node of the Summit supercomputer achieves a speed-up of almost 15× compared to running on central processing unit (CPU) only. Typical GPU kernels are memory-bound, achieving about 90% of peak. Our analysis shows that there is still room for improvement if we can refactor/fuse kernels to achieve higher arithmetic intensity. We also performed a weak parallel scalability study, which shows that the code runs well on a massively parallel system, but communication costs start becoming a significant bottleneck.

Germaschewski, K. (ORCID:0000000284956354)↗
DOE OSTI2021

GPU acceleration of rank-reduced coupled-cluster singles and doubles

Here, we have developed a graphical processing unit (GPU) accelerated implementation of our recently introduced rank-reduced coupled-cluster singles and doubles (RR-CCSD) method. RR-CCSD introduces a low-rank approximation of the doubles amplitudes. This is combined with a low-rank approximation of the electron repulsion integrals via Cholesky decomposition. The result of these two low-rank approximations is the replacement of the usual fourth-order CCSD tensors with products of second- and third-order tensors. In our implementation, only a single fourth-order tensor must be constructed as an intermediate during the solution of the amplitude equations. Owing in large part to the compression of the doubles amplitudes, the GPU-accelerated implementation shows excellent parallel efficiency (95% on eight GPUs). Our implementation can solve the RR-CCSD equations for up to 400 electrons and 1550 basis functions—roughly 50% larger than the largest canonical CCSD computations that have been performed on any hardware. In addition to increased scalability, the RR-CCSD computations are faster than the corresponding CCSD computations for all but the smallest molecules. We test the accuracy of RR-CCSD for a variety of chemical systems including up to 1000 basis functions and determine that accuracy to better than 0.1% error in the correlation energy can be achieved with roughly 95% compression of the ov space for the largest systems considered. We also demonstrate that conformational energies can be predicted to be within 0.1 kcal mol -1 with efficient compression applied to the wavefunction. Finally, we find that low-rank approximations of the CCSD doubles amplitudes used in the similarity transformation of the Hamiltonian prior to a conventional equation-of-motion CCSD computation will not introduce significant errors (on the order of a few hundredths of an electronvolt) into the resulting excitation energies.

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

Rank-reduced coupled-cluster. III. Tensor hypercontraction of the doubles amplitudes

In this work, we develop a quartic-scaling implementation of coupled-cluster singles and doubles (CCSD) based on low-rank tensor hypercontraction (THC) factorizations of both the electron repulsion integrals (ERIs) and the doubles amplitudes. This extends our rank-reduced (RR) coupled-cluster method to incorporate higher-order tensor factorizations. The THC factorization of the doubles amplitudes accounts for most of the gain in computational efficiency as it is sufficient, in conjunction with a Cholesky decomposition of the ERIs, to reduce the computational complexity of most contributions to the CCSD amplitude equations. Further THC factorization of the ERIs reduces the complexity of certain terms arising from nested commutators between the doubles excitation operator and the two-electron operator. We implement this new algorithm using graphical processing units and demonstrate that it enables CCSD calculations for molecules with 250 atoms and 2500 basis functions using a single computer node. Furthermore, we show that the new method computes correlation energies with comparable accuracy to the underlying RR-CCSD method.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗