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 577 records · Page 32
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

DGaaS: GPU as a Service on Distributed Computing System

In the rapidly evolving landscape of scientific computing, Graphics Processing Units (GPUs) have become indispensable for their unparalleled ability to handle parallel tasks in complex calculations, simulations, and data analysis. Their utility is further magnified in machine learning and AI applications, where they significantly accelerate model training and predictive analytics. Within this context, the Triton Inference Server emerges as a pivotal open-source tool, specializing in AI inferencing and optimizing GPU utilization across various platforms and frameworks. This paper presents an in-depth study on distributed High Throughput Computing (HTC), specifically focusing on the HTCondor framework and its resource provisioning tools, GlideinWMS and HEPCloud. These systems enable large-scale scientific experiments like CMS and DUNE to efficiently access and utilize vast computational resources. The paper explores the core architectural components of GlideinWMS, including jobs, user pools, and worker nodes, and discusses their integration with GPUs and the Triton server. The primary aim of this research is to develop a solution that optimizes GPU utilization by leveraging Glideins and containers. This approach allows computational jobs, particularly those involving AI models, to use GPUs only when essential, thereby facilitating efficient sharing of limited GPU resources. To validate this architecture, the study conducted three key tests involving custom scripts, container-based servers, and Triton server deployments. However, the study faces challenges, notably in locating the Triton server and ensuring secure remote access. To address these issues, future work will focus on developing a proxy mechanism and enhancing security protocols. In conclusion, this study offers a comprehensive roadmap for effective and efficient GPU utilization in distributed High Throughput Computing. It aims to contribute significantly to the scientific community by solving pressing problems and implementing robust solutions in collaboration with the GlideinWMS and HEPCloud teams. The research sets the stage for a more efficient, scalable, and cost-effective paradigm in scientific computing.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

FY17 MPACT Development Activities

Over the course of FY17, several improvements have been made to MPACT which effect the capability and performance of MPACT in several key areas. This report serves to document the capabilities developed at ORNL. There are four key areas of development that are highlighted here: • Externally driven source calculations to analyze reactor startup and the impact of secondary neutron sources. • Extensions to the Rhodium detector model to improve analysis of fixed core Rh detectors which are used in several cores. • Improved performance of PN scattering MOC sweeper to provide quicker reference solutions. • Implementation of an MOC kernel implemented on graphics processing units (GPU) to take advantage of advanced hardware. The next four sections discuss each of these features in detail. The final section discusses continuing work that is needed to improve these features.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2025

LAROMance Grade 91 Model Integration in NEML2

New reactor designs are targeting higher operating temperatures for increased thermal efficiency when compared to the current fleet of light water reactors. Designing structural components for these high temperature environments with reliable long-term operations requires material models that can accurately capture the deformation mechanisms active in these environments. The LAROMance surrogate material models are based on a database of mechanistic crystal plasticity simulations for high-temperature conditions. Inputs to the LAROMance models reflect the microstructural pedigree of the material, like dislocation densities and precipitate contents. Based on the evolution of these microstructural features, the LAROMance model provides the engineering scale constitutive model response. The LAROMance model was recently parameterized for Grade 91, a high temperature alloy. In the present work, the Grade 91 LAROMance model is implemented in the New Material Model Library, version 2 (NEML2). NEML2 provides a modular way to build material models from smaller blocks and was developed to vectorize the material update to efficiently run on modern computational architectures with graphics processing unit accelerators. NEML2 constitutive models can be used in simulations based on the multiphysics object-oriented simulation environment (MOOSE). This report provides details on the implementation of the Grade 91 LAROMance model in NEML2 and its verification of engineering scale finite element simulations in MOOSE.

42 - ENGINEERING↗
DOE OSTI2012

Accelerating the Computation of Detailed Chemical Reaction Kinetics for Simulating Combustion of Complex Fuels

Combustion of hydrocarbon fuels has been a very challenging scientific and engineering problem due to the complexity of turbulent flows and hydrocarbon reaction kinetics. There is an urgent need to develop an efficient modeling capability to accurately predict the combustion of complex fuels. Detailed chemical kinetic models for the surrogates of fuels such as gasoline, diesel and JP-8 consist of thousands of chemical species and Arrhenius reaction steps. Oxygenated fuels such as bio-fuels and heavier hydrocarbons, such as from newer fossil fuel sources, are expected to have a much more complex chemistry requiring increasingly larger chemical kinetic models. Such models are beyond current computational capability, except for homogeneous or partially stirred reactor type calculations. The advent of highly parallel multi-core processors and graphical processing units (GPUs) promises a steep increase in computational performance in the coming years. This paper will present a software framework that translates the detailed chemical kinetic models to high-performance code targeted for GPU accelerators.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2016

Large Eddy Simulation of a Supercritical Fuel Jet in Cross Flow using GPU-Acceleration

Fuel is routinely injected into combustion chambers at elevated pressures resulting in formation of supercritical fluid flow. The dispersion of fuel and the combustion processes are significantly different in high pressure conditions as the raditional liquid atomization and evaporation are replaced by diffusion-dominated mixing. Modeling supercritical flow is challenging due to the complex thermodynamics and transport properties of the fluids and in most cases high fidelity numerical simulations are prohibitively expensive. However, numerical investigations must incorporate the flow properties appropriate for the regime for accurate description of the physical processes involved. In this paper, we consider real-fluid properties and simulate the turbulent mixing of fuel jet in cross flow at high ambient pressure. In order to make efficient use of the computer resources and reduce the time to solution, without compromising the resolution requirements, the real-fluid property routines are accelerated via Graphics Processing Unit (GPU) computation. The numerical simulations performed at different resolutions suggest that the high fidelity computations with refined mesh are essential to capture the non-linear dynamics of vortex evolution downstream of the fuel jet and signifies the necessity for an accelerated code. The code is accelerated using the Kokkos C++ library and is demonstrated to be about 2 times faster than the non-accelerated version. Considerable computational cost associated with the thermodynamics routines is reduced in the accelerated version. The results also suggest the resource requirements to compute turbulent closures and the need for acceleration of the routines associated with the sub-grid scale closures.

Gottiparthi, Kalyana C.↗
DOE OSTI2019

An Adaptive Particle Tracking Algorithm for Lagrangian-Eulerian Simulations of Dispersed Multiphase Flows

The Lagrangian-Eulerian method is broadly used for simulations of dispersed multiphase flows by solving the continuous phase in the Eulerian framework while treating the dispersed phase as point particles in a Lagrangian framework. The accuracy of the Lagrangian-Eulerian method largely depends on the number of computational particles tracked for the Lagrangian phase. In this study, an adaptive Lagrangian particle tracking algorithm is proposed to balance the statistical error and the computational cost by dividing and merging the particles according to the local particle statistics in the computational domain. The computational particles are considered as weighted sampling points of the particle probability density functions (PDF) which represents a statistical equivalence of the Lagrangian phase. The algorithm is implemented as a part of a C++ library for Lagrangian particles, Grit, with performance portability to multi-core or many-core CPUs and Graphic Processing Unit (GPU) architectures. The accuracy of the proposed algorithm with two different schemes are studied through a test problem with analytical solutions for both the particle number density and momentum source term. The results are used to evaluate the proposed algorithm as well as to validate the parallel implementation.

Ge, Wenjun↗
DOE OSTI2023

Coupled cluster theory on modern heterogeneous supercomputers

This study examines the computational challenges in elucidating intricate chemical systems, particularly through ab-initio methodologies. This work highlights the Divide-Expand-Consolidate (DEC) approach for coupled cluster (CC) theory—a linear-scaling, massively parallel framework—as a viable solution. Detailed scrutiny of the DEC framework reveals its extensive applicability for large chemical systems, yet it also acknowledges inherent limitations. To mitigate these constraints, the cluster perturbation theory is presented as an effective remedy. Attention is then directed towards the CPS (D-3) model, explicitly derived from a CC singles parent and a doubles auxiliary excitation space, for computing excitation energies. The reviewed new algorithms for the CPS (D-3) method efficiently capitalize on multiple nodes and graphical processing units, expediting heavy tensor contractions. As a result, CPS (D-3) emerges as a scalable, rapid, and precise solution for computing molecular properties in large molecular systems, marking it an efficient contender to conventional CC models.

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

Enabling real-time adaptation of machine learning models at x-ray Free Electron Laser facilities with high-speed training optimized computational hardware

The emergence of novel computational hardware is enabling a new paradigm for rapid machine learning model training. For the Department of Energy’s major research facilities, this developing technology will enable a highly adaptive approach to experimental sciences. In this manuscript we present the per-epoch and end-to-end training times for an example of a streaming diagnostic that is planned for the upcoming high-repetition rate x-ray Free Electron Laser, the Linac Coherent Light Source-II. We explore the parameter space of batch size and data parallel training across multiple Graphics Processing Units and Reconfigurable Dataflow Units. We show the landscape of training times with a goal of full model retraining in under 15 min. Although a full from scratch retraining of a model may not be required in all cases, we nevertheless present an example of the application of emerging computational hardware for adapting machine learning models to changing environments in real-time, during streaming data acquisition, at the rates expected for the data fire hoses of accelerator-based user facilities.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

Adiabatic Quantum Computation Applied to Deep Learning Networks

Training deep learning networks is a difficult task due to computational complexity, and this is traditionally handled by simplifying network topology to enable parallel computation on graphical processing units (GPUs). However, the emergence of quantum devices allows reconsideration of complex topologies. We illustrate a particular network topology that can be trained to classify MNIST data (an image dataset of handwritten digits) and neutrino detection data using a restricted form of adiabatic quantum computation known as quantum annealing performed by a D-Wave processor. We provide a brief description of the hardware and how it solves Ising models, how we translate our data into the corresponding Ising models, and how we use available expanded topology options to explore potential performance improvements. Although we focus on the application of quantum annealing in this article, the work discussed here is just one of three approaches we explored as part of a larger project that considers alternative means for training deep learning networks. The other approaches involve using a high performance computing (HPC) environment to automatically find network topologies with good performance and using neuromorphic computing to find a low-power solution for training deep learning networks. Our results show that our quantum approach can find good network parameters in a reasonable time despite increased network topology complexity; that HPC can find good parameters for traditional, simplified network topologies; and that neuromorphic computers can use low power memristive hardware to represent complex topologies and parameters derived from other architecture choices.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Adaptive Fault Detection Based on Neural Networks and Multiple Sampling Points for Distribution Networks and Microgrids

Smart networks such as microgrid (MG) and active distribution networks (ADN) have been recently playing an important role in power system operation. The design and implementation of appropriate protection systems for such networks must be addressed, which imposes new technical challenges. This paper presents the implementation and validation aspects of an adaptive fault detection strategy based on neural networks (NNs) and multiple sampling points for ADN and MG. The solution is implemented on an edge device. Artificial NNs are used to derive a data-driven model that uses only local measurements to detect fault states of the network without the need for communication infrastructure. Multiple sampling points are used to derive a data-driven model, which allows the generalization considering the implementation in physical systems. The adaptive fault detector model is implemented on a Jetson Nano system, which is a single-board computer (SBC) with a small Graphic Processing Unit (GPU) intended to run machine learning loads at the edge. The proposed method is tested in a physical, real-life, low-voltage network located at Universidad del Norte, Colombia. This testing network is based on the IEEE-13 Node Test Feeder scaled down to 220 V. The validation in a simulation environment shows the accuracy and dependability above 99.6%, while the real-time tests show the accuracy and dependability of 95.5% and 100%, respectively. Without hard-to-derive parameters, the easy-to-implement embedded model highlights the potential for real-life applications.

42 ENGINEERING↗
DOE OSTI2016

THOR: A NEW AND FLEXIBLE GLOBAL CIRCULATION MODEL TO EXPLORE PLANETARY ATMOSPHERES

We have designed and developed, from scratch, a global circulation model (GCM) named THOR that solves the three-dimensional nonhydrostatic Euler equations. Our general approach lifts the commonly used assumptions of a shallow atmosphere and hydrostatic equilibrium. We solve the “pole problem” (where converging meridians on a sphere lead to increasingly smaller time steps near the poles) by implementing an icosahedral grid. Irregularities in the grid, which lead to grid imprinting, are smoothed using the “spring dynamics” technique. We validate our implementation of spring dynamics by examining calculations of the divergence and gradient of test functions. To prevent the computational time step from being bottlenecked by having to resolve sound waves, we implement a split-explicit method together with a horizontally explicit and vertically implicit integration. We validate our GCM by reproducing the Earth and hot-Jupiter-like benchmark tests. THOR was designed to run on graphics processing units (GPUs), which allows for physics modules (radiative transfer, clouds, chemistry) to be added in the future, and is part of the open-source Exoclimes Simulation Platform (www.exoclime.org).

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2016

FARGO3D: A NEW GPU-ORIENTED MHD CODE

We present the FARGO3D code, recently publicly released. It is a magnetohydrodynamics code developed with special emphasis on the physics of protoplanetary disks and planet–disk interactions, and parallelized with MPI. The hydrodynamics algorithms are based on finite-difference upwind, dimensionally split methods. The magnetohydrodynamics algorithms consist of the constrained transport method to preserve the divergence-free property of the magnetic field to machine accuracy, coupled to a method of characteristics for the evaluation of electromotive forces and Lorentz forces. Orbital advection is implemented, and an N-body solver is included to simulate planets or stars interacting with the gas. We present our implementation in detail and present a number of widely known tests for comparison purposes. One strength of FARGO3D is that it can run on either graphical processing units (GPUs) or central processing units (CPUs), achieving large speed-up with respect to CPU cores. We describe our implementation choices, which allow a user with no prior knowledge of GPU programming to develop new routines for CPUs, and have them translated automatically for GPUs.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2019

Excitation and Depletion of the Asteroid Belt in the Early Instability Scenario

Containing only a few percentages of the mass of the moon, the current asteroid belt is around three to four orders of magnitude smaller than its primordial mass inferred from disk models. Yet dynamical studies have shown that the asteroid belt could not have been depleted by more than about an order of magnitude over the past ∼4 Gyr. The remainder of the mass loss must have taken place during an earlier phase of the solar system’s evolution. An orbital instability in the outer solar system occurring during the process of terrestrial planet formation can reproduce the broad characteristics of the inner solar system. Here, we test the viability of this model within the constraints of the main belt’s low present-day mass and orbital structure. Although previous studies modeled asteroids as massless test particles because of limited computing power, our work uses graphics processing unit acceleration to model a fully self-gravitating asteroid belt. We find that depletion in the main belt is related to the giant planets’ exact evolution within the orbital instability. Simulations that produce the closest matches to the giant planets’ current orbits deplete the main belt by two to three orders of magnitude. These simulated asteroid belts are also good matches to the actual asteroid belt in terms of their radial mixing and broad orbital structure.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2024

Image Deconvolution and Point-spread Function Reconstruction with STARRED: A Wavelet-based Two-channel Method Optimized for Light-curve Extraction

We present starred, a point-spread function (PSF) reconstruction, two-channel deconvolution, and light-curve extraction method designed for high-precision photometric measurements in imaging time series. An improved resolution of the data is targeted rather than an infinite one, thereby minimizing deconvolution artifacts. In addition, starred performs a joint deconvolution of all available data, accounting for epoch-to-epoch variations of the PSF and decomposing the resulting deconvolved image into a point source and an extended source channel. The output is a high-signal-to-noise-ratio, high-resolution frame combining all data and the photometry of all point sources in the field of view as a function of time. Of note, starred also provides exquisite PSF models for each data frame. We showcase three applications of starred in the context of the imminent LSST survey and of JWST imaging: (i) the extraction of supernovae light curves and the scene representation of their host galaxy; (ii) the extraction of lensed quasar light curves for time-delay cosmography; and (iii) the measurement of the spectral energy distribution of globular clusters in the "Sparkler," a galaxy at redshift z = 1.378 strongly lensed by the galaxy cluster SMACS J0723.3-7327. starred is implemented in jax, leveraging automatic differentiation and graphics processing unit acceleration. This enables the rapid processing of large time-domain data sets, positioning the method as a powerful tool for extracting light curves from the multitude of lensed or unlensed variable and transient objects in the Rubin-LSST data, even when blended with intervening objects.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2017

An accurate and efficient algorithm for detection of radio bursts with an unknown dispersion measure, for single-dish telescopes and interferometers

Astronomical radio signals are subjected to phase dispersion while traveling through the interstellar medium. To optimally detect a short-duration signal within a frequency band, we have to precisely compensate for the unknown pulse dispersion, which is a computationally demanding task. We present the “fast dispersion measure transform” algorithm for optimal detection of such signals. Our algorithm has a low theoretical complexity of 2N{sub f}N{sub t}+N{sub t}N{sub Δ}log{sub 2}(N{sub f}), where N{sub f}, N{sub t}, and N{sub Δ} are the numbers of frequency bins, time bins, and dispersion measure bins, respectively. Unlike previously suggested fast algorithms, our algorithm conserves the sensitivity of brute-force dedispersion. Our tests indicate that this algorithm, running on a standard desktop computer and implemented in a high-level programming language, is already faster than the state-of-the-art dedispersion codes running on graphical processing units (GPUs). We also present a variant of the algorithm that can be efficiently implemented on GPUs. The latter algorithm’s computation and data-transport requirements are similar to those of a two-dimensional fast Fourier transform, indicating that incoherent dedispersion can now be considered a nonissue while planning future surveys. We further present a fast algorithm for sensitive detection of pulses shorter than the dispersive smearing limits of incoherent dedispersion. In typical cases, this algorithm is orders of magnitude faster than enumerating dispersion measures and coherently dedispersing by convolution. We analyze the computational complexity of pulsed signal searches by radio interferometers. We conclude that, using our suggested algorithms, maximally sensitive blind searches for dispersed pulses are feasible using existing facilities. We provide an implementation of these algorithms in Python and MATLAB.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2017

QSL Squasher: A Fast Quasi-separatrix Layer Map Calculator

Quasi-Separatrix Layers (QSLs) are a useful proxy for the locations where current sheets can develop in the solar corona, and give valuable information about the connectivity in complicated magnetic field configurations. However, calculating QSL maps, even for two-dimensional slices through three-dimensional models of coronal magnetic fields, is a non-trivial task, as it usually involves tracing out millions of magnetic field lines with immense precision. Thus, extending QSL calculations to three dimensions has rarely been done until now. In order to address this challenge, we present QSL Squasher—a public, open-source code, which is optimized for calculating QSL maps in both two and three dimensions on graphics processing units. The code achieves large processing speeds for three reasons, each of which results in an order-of-magnitude speed-up. (1) The code is parallelized using OpenCL. (2) The precision requirements for the QSL calculation are drastically reduced by using perturbation theory. (3) A new boundary detection criterion between quasi-connectivity domains is used, which quickly identifies possible QSL locations that need to be finely sampled by the code. That boundary detection criterion relies on finding the locations of abrupt field-line length changes, which we do by introducing a new Field-line Length Edge (FLEDGE) map. We find FLEDGE maps useful on their own as a quick-and-dirty substitute for QSL maps. QSL Squasher allows construction of high-resolution 3D FLEDGE maps in a matter of minutes, which is two orders of magnitude faster than calculating the corresponding 3D QSL maps. We include a sample of calculations done using QSL Squasher to demonstrate its capabilities as a QSL calculator, as well as to compare QSL and FLEDGE maps.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2022

GIGA-Lens: Fast Bayesian Inference for Strong Gravitational Lens Modeling

We present GIGA-Lens: a gradient-informed, GPU-accelerated Bayesian framework for modeling strong gravitational lensing systems, implemented in TensorFlow and JAX. The three components, optimization using multistart gradient descent, posterior covariance estimation with variational inference, and sampling via Hamiltonian Monte Carlo, all take advantage of gradient information through automatic differentiation and massive parallelization on graphics processing units (GPUs). We test our pipeline on a large set of simulated systems and demonstrate in detail its high level of performance. The average time to model a single system on four Nvidia A100 GPUs is 105 s. The robustness, speed, and scalability offered by this framework make it possible to model the large number of strong lenses found in current surveys and present a very promising prospect for the modeling of ${ \mathcal O }({10}^{5})$ lensing systems expected to be discovered in the era of the Vera C. Rubin Observatory, Euclid, and the Nancy Grace Roman Space Telescope.

79 ASTRONOMY AND ASTROPHYSICS↗