Search NASA⌕ Search

SEARCH · Search NASA

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 1,369 records · Page 76

Exactly and Easily Applying Experimental Boundary Conditions in Computational Structural Dynamics

Most experimental setups and environment specifications define acceleration loads on the component. However, Sierra Structural Dynamics cannot apply acceleration boundary conditions in modal transient analysis. Modal analysis of these systems and environments must be done through the application of a huge artificial force to a large fictitious point mass. Introducing a large mass into the analysis is a common source of numerical error. In this report we detail a mathematical procedure to directly apply acceleration boundary conditions in modal analyses without the requirement of adding a non-physical mass to the system. We prototype and demonstrate this procedure in Matlab and scope the work required to integrate this procedure into Sierra Structural Dynamics.

97 MATHEMATICS AND COMPUTING↗

Accelerating Time Integration for the Shallow Water Equations on the Sphere Using GPUs

The push towards larger and larger computational platforms has made it possible for climate simulations to resolve climate dynamics across multiple spatial and temporal scales. This direction in climate simulation has created a strong need to develop scalable time-stepping methods capable of accelerating throughput on high performance computing. This work details the recent advances in the implementation of implicit time stepping on a spectral element cube-sphere grid using graphical processing units (GPU) based machines. We demonstrate how solvers in the Trilinos project are interfaced with ACME and GPU kernels can significantly increase computational speed of the residual calculations in the implicit time stepping method for the shallow water equations on the sphere. We show the optimization gains and data structure reorganization that facilitates the performance improvements.

54 ENVIRONMENTAL SCIENCES↗

Iterative quantum-classical path integral with dynamically consistent state hopping

We investigate the convergence of iterative quantum-classical path integral calculations in sluggish environments strongly coupled to a quantum system. The number of classical trajectories, thus the computational cost, grows rapidly (exponentially, unless filtering techniques are employed) with the memory length included in the calculation. We argue that the choice of the (single) trajectory branch during the time preceding the memory interval can significantly affect the memory length required for convergence. At short times, the trajectory branch associated with the reactant state improves convergence by eliminating spurious memory. We also introduce an instantaneous population-based probabilistic scheme which introduces state-to-state hops in the retained pre-memory trajectory branch, and which is designed to choose primarily the trajectory branch associated with the reactant at early times, but to favor the product state more as the reaction progresses to completion. Test calculations show that the dynamically consistent state hopping scheme leads to accelerated convergence and a dramatic reduction of computational effort.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Density Tailored Capillary Discharge (Final Technical Report)

This final report describes efforts to validate a novel way to drive capillary discharges in order to achieve true matched guiding of a focused laser beam traveling along the capillary. Achieving such matched guiding is important for laser plasma accelerators, such as ones based upon laser wakefield acceleration. During this Phase II program, a computer model, which was specifically designed to predict the time-dependent behavior of the capillary discharge, was exercised to determine whether unconventional ways to drive the discharge may yield matched guiding. A custom power supply was built to drive the capillary discharge in the unconventional manner specified by the model. Experiments performed at STI Optronics seemed to support the predictions of this model; however, subsequent more accurate measurements performed at Lawrence Berkeley National Laboratory did not agree with the model predictions.

43 PARTICLE ACCELERATORS↗

Precise and Fast Computation of the Gravitational Field of a General Finite Body and Its Application to the Gravitational Study of Asteroid Eros

In order to obtain the gravitational field of a general finite body inside its Brillouin sphere, we developed a new method to compute the field accurately. First, the body is assumed to consist of some layers in a certain spherical polar coordinate system and the volume mass density of each layer is expanded as a Maclaurin series of the radial coordinate. Second, the line integral with respect to the radial coordinate is analytically evaluated in a closed form. Third, the resulting surface integrals are numerically integrated by the split quadrature method using the double exponential rule. Finally, the associated gravitational acceleration vector is obtained by numerically differentiating the numerically integrated potential. Numerical experiments confirmed that the new method is capable of computing the gravitational field independently of the location of the evaluation point, namely whether inside, on the surface of, or outside the body. It can also provide sufficiently precise field values, say of 14–15 digits for the potential and of 9–10 digits for the acceleration. Furthermore, its computational efficiency is better than that of the polyhedron approximation. This is because the computational error of the new method decreases much faster than that of the polyhedron models when the number of required transcendental function calls increases. As an application, we obtained the gravitational field of 433 Eros from its shape model expressed as the 24 × 24 spherical harmonic expansion by assuming homogeneity of the object.

79 ASTRONOMY AND ASTROPHYSICS↗

Convergence analysis of a CMFD method based on generalized equivalence theory

CMFD acceleration methods have been successful in reducing the computational burden for steady-state and transient reactor calculations. However, recent work on a complex coupled code BWR ATWS event has exposed possible issues with the stability of the CMFD method when standard CMFD methods are used. During the simulation of the ATWS boron injection event in the BWR, the PARCS code failed to converge with the existing CMFD method. A new CMFD method based on generalized equivalence theory was developed and the PARCS solution converged for the same ATWS event. This paper presents the new method and a detailed analytic and numerical convergence analysis. The results show that the new CMFD converges for all possible cross sections combinations anticipated in Light Water Reactor simulation and unlike existing CMFD methods, it is very robust even when the initial guess is far from final true solution. (authors)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

Convergence analysis of a CMFD method based on generalized equivalence theory

CMFD acceleration methods have been successful in reducing the computational burden for steady-state and transient reactor calculations. However, recent work on a complex coupled code BWR ATWS event has exposed possible issues with the stability of the CMFD method when standard CMFD methods are used. During the simulation of the ATWS boron injection event in the BWR, the PARCS code failed to converge with the existing CMFD method. A new CMFD method based on generalized equivalence theory was developed and the PARCS solution converged for the same ATWS event. This paper presents the new method and a detailed analytic and numerical convergence analysis. The results show that the new CMFD converges for all possible cross sections combinations anticipated in Light Water Reactor simulation and unlike existing CMFD methods, it is very robust even when the initial guess is far from final true solution. (authors)

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

Architectural Improvements and New Processing Tools for the Open XAL Online Model

The online model is the component of Open XAL providing accelerator modeling, simulation, and dynamic synchronization to live hardware. Significant architectural changes and feature additions have been recently made in two separate areas: 1) the managing and processing of simulation data, and 2) the modeling of RF cavities. Simulation data and data processing have been completely decoupled. A single class manages all simulation data while standard tools were developed for processing the simulation results. RF accelerating cavities are now modeled as composite structures where parameter and dynamics computations are distributed. The beam and hardware models both maintain their relative phase information, which allows for dynamic phase slip and elapsed time computation.

Accelerator Control↗

Computational Data-Driven Materials Discovery

Machine learning from large materials datasets enables accelerated materials discovery. Currently, the most accessible way to generate uniform, well curated, voluminous datasets is by the application of high-throughput first principles computations. In this work, we present the guiding principles of using computational data and machine learning to drive new materials discovery.

36 MATERIALS SCIENCE↗

IDEAL: Images Across Domains, Experiments, Algorithms and Learning

Research across science domains is increasingly reliant on image-centric data. Software tools are in high demand to uncover relevant, but hidden, information in digital images, such as those coming from faster next generation high-throughput imaging platforms. The challenge is to analyze the data torrent generated by the advanced instruments efficiently, and provide insights such as measurements for decision-making. In this article, we overview work performed by an interdisciplinary team of computational and materials scientists, aimed at designing software applications and coordinating research efforts connecting (1) emerging algorithms for dealing with large and complex datasets; (2) data analysis methods with emphasis in pattern recognition and machine learning; and (3) advances in evolving computer architectures. Engineering tools around these efforts accelerate the analyses of image-based recordings, improve reusability and reproducibility, scale scientific procedures by reducing time between experiments, increase efficiency, and open opportunities for more users of the imaging facilities. This paper describes our algorithms and software tools, showing results across image scales, demonstrating how our framework plays a role in improving image understanding for quality control of existent materials and discovery of new compounds.

97 MATHEMATICS AND COMPUTING↗

Uncertainty quantification for Multiphase-CFD simulations of bubbly flows: a machine learning-based Bayesian approach supported by high-resolution experiments

In this paper, we developed a machine learning-based Bayesian approach to inversely quantify and reduce the uncertainties of multiphase computational fluid dynamics (MCFD) simulations for bubbly flows. The proposed approach is supported by high-resolution two-phase flow measurements, including those by double-sensor conductivity probes, high-speed imaging, and particle image velocimetry. Local distributions of key physical quantities of interest (QoIs), including the void fraction and phasic velocities, are obtained to support the Bayesian inference. In the process, the epistemic uncertainties of the closure relations are inversely quantified while the aleatory uncertainties from stochastic fluctuations of the system are evaluated based on experimental uncertainty analysis. The combined uncertainties are then propagated through the MCFD solver to obtain uncertainties of the QoIs, based on which probability-boxes are constructed for validation. The proposed approach relies on three machine learning methods: feedforward neural networks and principal component analysis for surrogate modeling, and Gaussian processes for model form uncertainty modeling. The whole process is implemented within the framework of an open-source deep learning library PyTorch with graphics processing unit (GPU) acceleration, thus ensuring the efficiency of the computation. The results demonstrate that with the support of high-resolution data, the uncertainties of MCFD simulations can be significantly reduced. The proposed approach has the potential for other applications that involve numerical models with empirical parameters.

42 ENGINEERING↗

Job Management with mpi_jm

About this book: This book constitutes the refereed post-conference proceedings of 13 workshops held at the 33rd International ISC High Performance 2018 Conference, in Frankfurt, Germany, in June 2018: HPC I/O in the Data Center, HPC-IODC 2018; Workshop on Performance and Scalability of Storage Systems, WOPSSS 2018; 13th Workshop on Virtualization in High­-Performance Cloud Computing, VHPC 2018; Third International Workshop on In Situ Visualization, WOIV 2018; 4th International Workshop on Communication Architectures for HPC, Big Data, Deep Learning and Clouds at Extreme Scale, ExaComm 2018; International Workshop on OpenPOWER for HPC, IWOPH 2018; IXPUG Workshop: Many-Core Computing on Intel Processors; Workshop on Sustainable Ultrascale Computing Systems; Approximate and Transprecision Computing on Emerging Technologies, ATCET 2018; First Workshop on the Convergence of Large-Scale Simulation and Artificial Intelligence; Third Workshop for Open Source Supercomputing, OpenSuCo 2018; First Workshop on Interactive High-Performance Computing; Workshop on Performance Portable Programming Models for Accelerators, P^3MA 2018. The 53 full papers included in this volume were carefully reviewed and selected from 80 submissions. They cover all aspects of research, development, and application of large-scale, high performance experimental and commercial systems. Topics include HPC computer architecture and hardware; programming models, system software, and applications; solutions for heterogeneity, reliability, power efficiency of systems; virtualization and containerized environments; big data and cloud computing; and artificial intelligence.

Berkowitz, Evan↗

Acceleration of Graph Neural Network-Based Prediction Models in Chemistry via Co-Design Optimization on Intelligence Processing Units

Atomic structure prediction and associated property calculations are the bedrock of chemical physics. Since high-fidelity ab initio modeling techniques for computing the structure and properties can be prohibitively expensive, this motivates the development of machine-learning (ML) models that make these predictions more efficiently. Training graph neural networks over large atomistic databases introduces unique computational challenges such as the need to process millions of small graphs with variable size and support communication patterns that are distinct from learning over large graphs such as social networks. We demonstrate a novel hardware-software co-design approach to scale up the training of atomistic graph neural networks (GNN) for structure and property prediction. First, to eliminate redundant computation and memory associated with alternative padding techniques and to improve throughput via minimizing communication, we formulate the effective coalescing of the batches of variable-size atomistic graphs as the bin packing problem and introduce a hardware-agnostic algorithm to pack these batches. In addition, we propose hardware-specific optimizations including a planner and vectorization for the gather-scatter operations targeted for Graphcore’s Intelligence Processing Unit (IPU), as well as model-specific optimizations such as merged communication collectives and optimized softplus. Putting these all together, we demonstrate the effectiveness of the proposed co-design approach by providing an implementation of a well-established atomistic GNN on the Graphcore IPUs. We evaluate the training performance on multiple atomistic graph databases with varying degrees of graph counts, sizes and sparsity. Here, we demonstrate that such a co-design approach can reduce the training time of atomistic GNNs and can improve the performance by up to 1.5× compared to the baseline implementation of the model on the IPUs. Additionally, we compare our IPU implementation with a Nvidia GPU-based implementation and show that our atomistic GNN implementation on the IPUs can run 1.8× faster on average compared to the execution time on the GPUs.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

INTEGRATED WORKFLOW MANAGEMENT FOR PARTICLE ACCELERATOR SIMULATION

Supercomputing systems are used for a wide range of computationally demanding tasks in many fields of science and engineering. They play a key role in numerical simulation, in which mathematical models are computed in order to simulate the behavior of physical systems. Scientists and engineers that use supercomputers for numerical simulation often have their productivity limited by the need to manually organize and manage extremely large amounts of data that are often produced and consumed by the software programs run on these systems. Recognizing these limitations, Kitware Inc. (Clifton Park, NY) and SLAC National Accelerator Laboratory (Menlo Park, CA) are developing an advanced software platform that can reduce the cognitive overhead required by knowledge workers when using supercomputers for numerical simulation. Phase I of the project is complete and includes the development of new capabilities for organizing simulation project files, improvements to the user interface and overall usability, and deployment of a “middle tier” server to sit between user desktop machines and supercomputers to offload much of the data management workload. The project also developed prototype software for executing sequences of numerical simulations, and a prototype for migrating supercomputing software to cloud-based computing systems to provide a potential alternative to supercomputers with different logistical and price-to-performance tradeoffs.

Tourtellott, John↗

Accelerated Dimension-Independent Adaptive Metropolis

This work describes improvements from algorithmic and architectural means to black-box Bayesian inference over high-dimensional parameter spaces. The well-known adaptive Metropolis (AM) algorithm [33] is extended herein to scale asymptotically uniformly with respect to the underlying parameter dimension for Gaussian targets, by respecting the variance of the target. The resulting algorithm, referred to as the dimension-independent adaptive Metropolis (DIAM) algorithm, also shows improved performance with respect to adaptive Metropolis on non-Gaussian targets. This algorithm is further improved, and the possibility of probing high-dimensional (with dimension d 1000) targets is enabled, via GPU-accelerated numerical libraries and periodically synchronized concurrent chains (justi ed a posteriori). Asymptotically in dimension, this GPU implementation exhibits a factor of four improvement versus a competitive CPU-based Intel MKL parallel version alone. Strong scaling to concurrent chains is exhibited, through a combination of longer time per sample batch (weak scaling) and yet fewer necessary samples to convergence. The algorithm performance is illustrated on several Gaussian and non-Gaussian target examples, in which the dimension may be in excess of one thousand.

97 MATHEMATICS AND COMPUTING↗

LLM-Inference-Bench: Inference Benchmarking of Large Language Models on AI Accelerators

Large Language Models (LLMs) have propelled groundbreaking advancements across several domains and are commonly used for text generation applications. However, the computational demands of these complex models pose significant challenges, requiring efficient hardware acceleration. Benchmarking the performance of LLMs across diverse hardware platforms is crucial to understanding their scalability and throughput characteristics. We introduce LLM-Inference-Bench, a comprehensive benchmarking suite to evaluate the hardware inference performance of LLMs. We thoroughly analyze diverse hardware platforms, including GPUs from Nvidia and AMD and specialized AI accelerators, Intel Habana and SambaNova. Our evaluation includes several LLM inference frameworks and models from LLaMA, Mistral, and Qwen families with 7B and 70B parameters. Our benchmarking results reveal the strengths and limitations of various models, hardware platforms, and inference frameworks. We provide an interactive dashboard to help identify configurations for optimal performance for a given hardware platform.

Chitty-Venkata, Krishna Teja↗

PARTICLE ACCELERATION AT QUASI-PARALLEL SHOCK WAVES: THEORY AND OBSERVATIONS AT 1 AU

In this paper, we describe a theoretical model for accelerating an arbitrary upstream particle distribution. Only those particles that exceed a prescribed injection energy, E{sub inj}, are accelerated via the diffusive shock acceleration (DSA) mechanism, also known as first-order Fermi acceleration. We identify a set of quasi-parallel shocks at 1 AU and use the observed solar wind particle distribution information to construct our upstream distribution, which is then accelerated diffusively at the shock, assuming the observed shock parameters. The injection energy for particles to be accelerated diffusively at a quasi-parallel shock is discussed theoretically. By using the observed upstream solar wind distribution function and the observed shock parameters, we can compute the injection energy that matches the observed downstream accelerated particle spectrum. Like the previous studies of van Nes et al., Lario et al., and Ho et al., this analysis focuses on the acceleration of protons only via the first-order Fermi acceleration mechanism. However, our primary focus is on quasi-parallel shocks and the injection mechanism in the context of DSA with a background thermal solar wind modeled as a Maxwellian or kappa distribution. Our approach allows for a direct test of injection at interplanetary shocks. It has been proposed that an additional seed population of energetic particles is needed to explain the accelerated particle distribution downstream of quasi-parallel shocks. This conclusion is based typically on studies that address the acceleration of heavy ions primarily and do not characterize the injection of protons alone using the DSA mechanism. Through comparisons of Maxwellian and kappa upstream distributions, we find that DSA with injection directly from a thermal Maxwellian distribution, or weak departures therefrom, for protons is responsible for energetic solar particle events associated with quasi-parallel shocks.

79 ASTRONOMY AND ASTROPHYSICS↗