Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “accelerated computing”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

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

At least 829 records · Page 46
All sourcesNASA NTRSDOE OSTI
NASA NTRS2019

CFD Jet Mixing Model Validation Against Zero-Boil-Off Tank (ZBOT) Microgravity Experiment

The Zero-Boil-Off Tank (ZBOT) experiments are a series of small scale two-phase fluids physics experiments conducted aboard the International Space Station (ISS) in 2017. A goal of these experiments is to study various fundamental fluid flow, heat and mass transport, and phase change phenomena that control storage tank pressurization and pressure control in microgravity. In this paper, results of a state-of-the-art two-phase CFD (Computational Fluid Dynamics) storage tank model are presented and validated against jet mixing data provided by microgravity ZBOT experiments. Jet-ullage interaction was studied in the experiment for different jet speeds representing different flow regimes, and tank fill levels. Effects of various physical parameters, such as, jet angle, gravitational acceleration and surface tension value were investigated. Computational parameters, such as turbulence and surface tension models and interface capturing scheme, affecting jet-ullage interaction were also studied. Computational results were compared to experimental results for model validation.

Turbulence↗
DOE OSTI2024

AXI4MLIR: User-Driven Automatic Host Code Generation for Custom AXI-Based Accelerators

Tensor algebra operations represent an important class of algorithms used across many applications, including machine learning, scientific computing, and data analytics. As a result, the efficient generation of custom accelerators for tensor operations has received increased attention. Previous efforts have produced automated tools enabling users to prototype and explore optimized accelerators. However, little effort has been focused on the host-accelerator interaction in these tools. Efficient use of hardware accelerators requires knowledge about the accelerator's capabilities (operations, data formats, and opcode support), the host CPU microarchitecture (e.g., memory hierarchy), the host-accelerator interface, and the application's features (which code regions should be mapped onto an accelerator). Manually rewriting the original applications to facilitate improved custom accelerator mapping is an error-prone and time-consuming endeavor. To cope with this, we propose AXI4MLIR, a new framework to automatically generate and optimize the communication between the host CPU and arbitrary accelerators that implement linear algebra algorithms. AXI4MLIR extends the MLIR compiler framework to automatically generate efficient host-accelerator driver code for accelerators with AXI-based interfaces. Our compiler extensions enable automatic driver code generation while carefully considering the host's memory hierarchy and target accelerator features. To demonstrate the flexibility and utility of AXI4MLIR, we test it with diverse use cases that include different types of accelerators, tiling scenarios, and dataflow schemes. We compare our experimental results to manual implementations of host-accelerator driver code and find that our approach can reduce CPU cache references by 56% and deliver up to a 1.65x speedup.

Bohm Agostini, Nicolas↗
DOE OSTI2023

Modeling Codes Used for LAMP

LAMP team is planning on using multiple computer codes as tools for modeling beam dynamics, electromagnetics, and plasma environment in ion sources. Below is a description of the computer tools we have been using or planning to use in the future. Most of the codes chosen for the LAMP project are capable to model most of the beam dynamics in LAMP. A judicious choice of the appropriate code to use will be based on speed and fidelity of the modeling with faster calculations produced by simplifying some features.

43 PARTICLE ACCELERATORS↗
DOE OSTI2022

On the integration of molecular dynamics, data science, and experiments for studying solvent effects on catalysis

Computational workflows that combine molecular dynamics (MD) simulations and emerging data-centric (DC) methods can accelerate the screening and analysis of solvent systems experimentally and computationally. Here, MD simulations provide atomic positions and velocities of reactant, solvent, and catalyst materials that can be manipulated into data representations that in turn can be used by DC techniques to conduct predictive modeling, feature extraction, and experimental design. For liquid-phase catalytic applications, emerging DC techniques such as Convolutional and Graph Neural Networks (CNN/GNN), Topological Data Analysis (TDA), and Active Learning (AL) can leverage MD and experimental data to quickly predict solvent effects on reaction outcomes. For instance, in recent studies, 3D solvent environments obtained with MD have been exploited by CNNs to predict experimental reaction rates for homogeneous acid-catalyzed lignocellulosic processes. In this perspective, we discuss basic principles of DC methods and how these can be combined with MD to enable high-throughput screening of solvent selection for diverse catalysis applications.

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

Analysis and mitigation of parasitic resistance effects for analog in-memory neural network acceleration

To support the increasing demands for efficient deep neural network processing, accelerators based on analog in-memory computation of matrix multiplication have recently gained significant attention for reducing the energy of neural network inference. However, analog processing within memory arrays must contend with the issue of parasitic voltage drops across the metal interconnects, which distort the results of the computation and limit the array size. This work analyzes how parasitic resistance affects the end-to-end inference accuracy of state-of-the-art convolutional neural networks, and comprehensively studies how various design decisions at the device, circuit, architecture, and algorithm levels affect the system's sensitivity to parasitic resistance effects. Here, a set of guidelines are provided for how to design analog accelerator hardware that is intrinsically robust to parasitic resistance, without any explicit compensation or re-training of the network parameters.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1984

Computational aerodynamics and artificial intelligence

The general principles of artificial intelligence are reviewed and speculations are made concerning how knowledge based systems can accelerate the process of acquiring new knowledge in aerodynamics, how computational fluid dynamics may use expert systems, and how expert systems may speed the design and development process. In addition, the anatomy of an idealized expert system called AERODYNAMICIST is discussed. Resource requirements for using artificial intelligence in computational fluid dynamics and aerodynamics are examined. Three main conclusions are presented. First, there are two related aspects of computational aerodynamics: reasoning and calculating. Second, a substantial portion of reasoning can be achieved with artificial intelligence. It offers the opportunity of using computers as reasoning machines to set the stage for efficient calculating. Third, expert systems are likely to be new assets of institutions involved in aeronautics for various tasks of computational aerodynamics.

Mehta, U. B.↗
NASA NTRS1991

The plasma physics of shock acceleration

The history and theory of shock acceleration is reviewed, paying particular attention to theories of parallel shocks which include the backreaction of accelerated particles on the shock structure. The work that computer simulations, both plasma and Monte Carlo, are playing in revealing how thermal ions interact with shocks and how particle acceleration appears to be an inevitable and necessary part of the basic plasma physics that governs collisionless shocks is discussed. Some of the outstanding problems that still confront theorists and observers in this field are described.

Jones, Frank C.↗
DOE OSTI2023

HipBone: A performance-portable graphics processing unit-accelerated C++ version of the NekBone benchmark

We present hipBone, an open-source performance-portable proxy application for the Nek5000 (and NekRS) computational fluid dynamics applications. HipBone is a fully GPU-accelerated C++ implementation of the original NekBone CPU proxy application with several novel algorithmic and implementation improvements which optimize its performance on modern fine-grain parallel GPU accelerators. Our optimizations include a conversion to store the degrees of freedom of the problem in assembled form in order to reduce the amount of data moved during the main iteration and a portable implementation of the main Poisson operator kernel. We demonstrate near-roofline performance of the operator kernel on three different modern GPU accelerators from two different vendors. We present a novel algorithm for splitting the application of the Poisson operator on GPUs which aggressively hides MPI communication required for both halo exchange and assembly. Our implementation of nearest-neighbor MPI communication then leverages several different routing algorithms and GPU-Direct RDMA capabilities, when available, which improves scalability of the benchmark. We demonstrate the performance of hipBone on three different clusters housed at Oak Ridge National Laboratory, namely, the Summit supercomputer and the Frontier early-access clusters, Spock and Crusher. Our tests demonstrate both portability across different clusters and very good scaling efficiency, especially on large problems.

Computer Science↗
DOE OSTI2021

Machine-learning accelerated geometry optimization in molecular simulation

Geometry optimization is an important part of both computational materials and surface science because it is the path to finding ground state atomic structures and reaction pathways. These properties are used in the estimation of thermodynamic and kinetic properties of molecular and crystal structures. This process is slow at the quantum level of theory because it involves an iterative calculation of forces using quantum chemical codes such as density functional theory (DFT), which are computationally expensive and which limit the speed of the optimization algorithms. It would be highly advantageous to accelerate this process because then one could do either the same amount of work in less time or more work in the same time. Here, we provide a neural network (NN) ensemble based active learning method to accelerate the local geometry optimization for multiple configurations simultaneously. We illustrate the acceleration on several case studies including bare metal surfaces, surfaces with adsorbates, and nudged elastic band for two reactions. In all cases, the accelerated method requires fewer DFT calculations than the standard method. In addition, we provide an Atomic Simulation Environment (ASE)-optimizer Python package to make the usage of the NN ensemble active learning for geometry optimization easier.

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

Direct evidence of microstructure dependence of magnetic flux trapping in niobium

Abstract Elemental type-II superconducting niobium is the material of choice for superconducting radiofrequency cavities used in modern particle accelerators, light sources, detectors, sensors, and quantum computing architecture. An essential challenge to increasing energy efficiency in rf applications is the power dissipation due to residual magnetic field that is trapped during the cool down process due to incomplete magnetic field expulsion. New SRF cavity processing recipes that use surface doping techniques have significantly increased their cryogenic efficiency. However, the performance of SRF Nb accelerators still shows vulnerability to a trapped magnetic field. In this manuscript, we report the observation of a direct link between flux trapping and incomplete flux expulsion with spatial variations in microstructure within the niobium. Fine-grain recrystallized microstructure with an average grain size of 10–50 µm leads to flux trapping even with a lack of dislocation structures in grain interiors. Larger grain sizes beyond 100–400 µm do not lead to preferential flux trapping, as observed directly by magneto-optical imaging. While local magnetic flux variations imaged by magneto-optics provide clarity on a microstructure level, bulk variations are also indicated by variations in pinning force curves with sequential heat treatment studies. The key results indicate that complete control of the niobium microstructure will help produce higher performance superconducting resonators with reduced rf losses 1 related to the magnetic flux trapping.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2021

Low Power Hardware-In-The-Loop Neuromorphic Training Accelerator

The training process for spiking neural networks can be very computationally intensive. Approaches such as evolutionary algorithms may require evaluating thousands or millions of candidate solutions. In this work, we propose using neuromorphic cores implemented on a Xilinx Zynq system on chip to accelerate and improve the energy efficiency of the evaluation step of an evolutionary training approach. We demonstrate this can significantly reduce the required energy to evolve a network with some cases showing greater than 10 times improvement as compared to a CPU-only system.

Mitchell, Parker↗
DOE OSTI2026

Automated and Accelerated Continuum Model Development for Electrochemical Systems (Abbreviated Report)

Despite the availability of computational resources and advancements in numerical computing capabilities, the multiscale models core to understanding, predicting the behaviors of, and designing energy and environmental systems involving porous media are still 1.) developed through by-hand derivations and 2.) limited by many methodological assumptions employed during model derivation. As a result, the advancement of effective media models for engineering DOE mission-critical systems (e.g., batteries, flow batteries, electrolyzers, geothermal systems, subsurface chemical storage systems, etc.) is slow (i.e., it takes years for models to traverse from stages of “development” to “practical utilization”), hindering our ability to effectively optimize such systems and stay at the cutting-edge of the energy frontier. In this work, we aimed to address these limitations by 1.) automating and accelerating multiscale model derivation via symbolic computing and 2.) develop a novel multiscale modeling methodology for flow and transport through porous media that avoids the typical assumptions hindering previous models. As a result of our efforts, we 1.) developed a hybrid symbolic-numeric code called Fouriera for fully-automating the implementation of multiphysical and phase-field models via the Fourier spectral method for materials science research, and 2.) advanced a multiscale modeling methodology called The Method of Finite Averages that rigorously predicts the behaviors of flow and transport through heterogeneous porous media under the influence of non-local effects and strong advection. Ultimately, these deliverables provide strong foundations from which further efforts can advance multiscale modeling tools and capabilities that do not intrinsically rely on 1.) the speed and mathematical capabilities of humans, nor 2.) the methodological assumptions limiting current models.

36 MATERIALS SCIENCE↗
NASA NTRS1984

Computational aerodynamics and artificial intelligence

Some aspects of artificial intelligence are considered and questions are speculated on, including how knowledge-based systems can accelerate the process of acquiring new knowledge in aerodynamics, how computational fluid dynamics may use 'expert' systems and how expert systems may speed the design and development process. The anatomy of an idealized expert system called AERODYNAMICIST is discussed. Resource requirements are examined for using artificial intelligence in computational fluid dynamics and aerodynamics. Considering two of the essentials of computational aerodynamics - reasoniing and calculating - it is believed that a substantial part of the reasoning can be achieved with artificial intelligence, with computers being used as reasoning machines to set the stage for calculating. Expert systems will probably be new assets of institutions involved in aeronautics for various tasks of computational aerodynamics.

Kutler, P.↗
NASA NTRS1993

Simulations of particle acceleration in parallel shocks: Direct comparison between Monte Carlo and one-dimensional hybrid codes

We have made a direct comparison between two different computer simulations of a plane, parallel, collisionless shock including particle acceleration to energies typical of those of diffuse ions observed at the earth bow shock. Despite the fact that the one-dimensional hybrid and Monte Carlo techniques employ entirely different algorithms, they give surprisingly close agreement in the overall shapes of the complete distribution functions for protons as well as heavier ions. Both methods show that energetic ions emerge smoothly from the background thermal plasma with approximately the same relative injection rate and that the fraction of the incoming plasma's energy flux that is converted into downstream enthalpy flux of the accelerated population (i.e., the acceleration efficiency) is similar in the two cases. The fraction of the downstream proton distribution made up of superthermal particles is quite large, with at least 10% of the energy flux going into protons with energies above 10 keV. In addition, an upstream precursor, produced by backstreaming energetic particles, is present in both shocks, although the Monte Carlo precursor is considerably longer than that produced in the hybrid shock. These results offer convincing evidence that, at least in these ways, the two simulations are consistent in their description of parallel shock structure and particle acceleration, and they lay the groundwork for development of shock models employing a combination of both methods.

Ellison, Donald C.↗
DOE OSTI2020

CoSim: A Simulator for Co-Scheduling of Batch and On-Demand Jobs in HPC Datacenters

The increasing scale and complexity of scientific applications are rapidly transforming the ecosystem of tools, methods, and workflows adopted by the high-performance computing (HPC) community. Big data analytics and deep learning are gaining traction as essential components in this ecosystem in a variety of scenarios, such as, steering of experimental instruments , acceleration of high-fidelity simulations through surrogate computations, and guided ensemble searches. In this context, the batch job model traditionally adopted by the supercomputing in-frastructures needs to be complemented with support to schedule opportunistic on-demand analytics jobs, leading to the problem of efficient preemption of batch jobs with minimum loss of progress. In this paper, we design and implement a simulator, CoSim, that enables on-the-fly analysis of the trade-offs arising between delaying the start of opportunistic on-demand jobs, which leads to longer analytics latency, and loss of progress due to preemption of batch jobs, which is necessary to make room for on-demand jobs. To this end, we propose an algorithm based on dynamic programming with predictable performance and scalability that enables supercomputing infrastructure schedulers to analyze the aforementioned trade-off and take decisions in near real-time. Compared with other state-of-art approaches using traces of the Theta pre-Exascale machine, our approach is capable of finding the optimal solution, while achieving high performance and scalability.

batch job preemption↗
DOE OSTI2022

Inference-Optimized AI and High Performance Computing for Gravitational Wave Detection at Scale

We introduce an ensemble of artificial intelligence models for gravitational wave detection that we trained in the Summit supercomputer using 32 nodes, equivalent to 192 NVIDIA V100 GPUs, within 2 h. Once fully trained, we optimized these models for accelerated inference using NVIDIA TensorRT. We deployed our inference-optimized AI ensemble in the ThetaGPU supercomputer at Argonne Leadership Computer Facility to conduct distributed inference. Using the entire ThetaGPU supercomputer, consisting of 20 nodes each of which has 8 NVIDIA A100 Tensor Core GPUs and 2 AMD Rome CPUs, our NVIDIA TensorRT-optimized AI ensemble processed an entire month of advanced LIGO data (including Hanford and Livingston data streams) within 50 s. Our inference-optimized AI ensemble retains the same sensitivity of traditional AI models, namely, it identifies all known binary black hole mergers previously identified in this advanced LIGO dataset and reports no misclassifications, while also providing a 3X inference speedup compared to traditional artificial intelligence models. We used time slides to quantify the performance of our AI ensemble to process up to 5 years worth of advanced LIGO data. In this synthetically enhanced dataset, our AI ensemble reports an average of one misclassification for every month of searched advanced LIGO data. We also present the receiver operating characteristic curve of our AI ensemble using this 5 year long advanced LIGO dataset. This approach provides the required tools to conduct accelerated, AI-driven gravitational wave detection at scale.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Graph Neural Network-based Tracking as a Service

Recent studies have shown promising results for track finding in dense environments using Graph Neural Network (GNN)-based algorithms. However, GNN-based track finding is computationally slow on CPUs, necessitating the use of coprocessors to accelerate the inference time. Additionally, the large input graph size demands a large device memory for efficient computation, a requirement not met by all computing facilities used for particle physics experiments, particularly those lacking advanced GPUs. Furthermore, deploying the GNN-based track-finding algorithm in a production environment requires the installation of all dependent software packages, exclusively utilized by this algorithm. These computing challenges must be addressed for the successful implementation of GNN-based track-finding algorithm into production settings. In response, we introduce a ``GNN-based tracking as a service'' approach, incorporating a custom backend within the NVIDIA Triton inference server to facilitate GNN-based tracking. This paper presents the performance of this approach using the Perlmutter supercomputer at NERSC.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
NASA NTRS2017

Damage Based Analysis (DBA) - Theory, Derivation and Practical Application Using Both an Acceleration and Pseudo Velocity Approach

The objective of this presentation is to give a brief overview of the theory behind the (DBA) method, an overview of the derivation and a practical application of the theory using the Python computer language. The Theory and Derivation will use both Acceleration and Pseudo Velocity methods to derive a series of equations for processing by Python. We will take the results and compare both Acceleration and Pseudo Velocity methods and discuss implementation of the Python functions. Also, we will discuss the efficiency of the methods and the amount of computer time required for the solution. In conclusion, (DBA) offers a powerful method to evaluate the amount of energy imparted into a system in the form of both Amplitude and Duration during qualification testing and flight environments. Many forms of steady state and transient vibratory motion can be characterized using this technique. (DBA) provides a more robust alternative to traditional methods such Power Spectral Density (PSD) using a maximax approach.

Structural Dynamics↗