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 829 records · Page 46

An introduction to neuromorphic computing and its potential impact for unattended ground sensors

Neuromorphic computers are hardware systems that mimic the brain’s computational process phenomenology. This is in contrast to neural network accelerators, such as the Google TPU or the Intel Neural Compute Stick, which seek to accelerate the fundamental computation and data flows of neural network models used in the field of machine learning. Neuromorphic computers emulate the integrate and fire neuron dynamics of the brain to achieve a spiking communication architecture for computation. While neural networks are brain-inspired, they drastically oversimplify the brain’s computation model. Neuromorphic architectures are closer to the true computation model of the brain (albeit, still simplified). Neuromorphic computing models herald a 1000x power improvement over conventional CPU architectures. Sandia National Labs is a major contributor to the research community on neuromorphic systems by performing design analysis, evaluation, and algorithm development for neuromorphic computers. Space-based remote sensing development has been a focused target of funding for exploratory research into neuromorphic systems for their potential advantage in that program area; SNL has led some of these efforts. Recently, neuromorphic application evaluation has reached the NA-22 program area. This same exploratory research and algorithm development should penetrate the unattended ground sensor space for SNL’s mission partners and program areas. Neuromorphic computing paradigms offer a distinct advantage for the SWaP-constrained embedded systems of our diverse sponsor-driven program areas.

97 MATHEMATICS AND COMPUTING↗

ATHENA: Analytical Tool for Heterogeneous Neuromorphic Architectures

The ASC program seeks to use machine learning to improve efficiencies in its stockpile stewardship mission. Moreover, there is a growing market for technologies dedicated to accelerating AI workloads. Many of these emerging architectures promise to provide savings in energy efficiency, area, and latency when compared to traditional CPUs for these types of applications — neuromorphic analog and digital technologies provide both low-power and configurable acceleration of challenging artificial intelligence (AI) algorithms. If designed into a heterogeneous system with other accelerators and conventional compute nodes, these technologies have the potential to augment the capabilities of traditional High Performance Computing (HPC) platforms [5]. This expanded computation space requires not only a new approach to physics simulation, but the ability to evaluate and analyze next-generation architectures specialized for AI/ML workloads in both traditional HPC and embedded ND applications. Developing this capability will enable ASC to understand how this hardware performs in both HPC and ND environments, improve our ability to port our applications, guide the development of computing hardware, and inform vendor interactions, leading them toward solutions that address ASC’s unique requirements.

97 MATHEMATICS AND COMPUTING↗

Modeling and Simulation Development Pathways to Accelerating KP-FHR Licensing (Final Report)

This project assembles a strong U.S. industry and national laboratory team to complete scope of work. Kairos Power (KP), headquartered in Alameda, CA, is the leader of this effort and has built an internal team of highly competent engineers and managers with extensive combined experience in nuclear power, conventional power, product development, and licensing. INL, ANL, and LANL bring unique capabilities in advanced reactor R&D and licensing. The project funding source is the result of FOA No. 0001817, U. S. Industry Opportunities for Advanced Nuclear Technology Development. There has been on-going work and this Access Cooperative Research and Development Agreement (CRADA) will cover the remaining work scope of FOA 0001817. KP is implementing innovative strategies that can reduce the cost and accelerate the initial demonstration of the Kairos Power Fluoride-salt-cooled, High-temperature Reactor (KP-FHR) to meet the needs of the U.S. electricity market by 2030. Licensing of the KP-FHR could be significantly accelerated using advanced computing methods with sufficient predictive capabilities to be able to extrapolate potential response of the structure in different scenarios. However, currently used computational methods heavily rely on empirical fits and cannot be used for extrapolation. The scope focuses on improving the modeling capability in the NEAMS Grizzly structural mechanics code and the NEAMS BISON fuel performance code and the NEAMS SAM systems analysis code. This work leverages the expertise and know-how gathered in three DOE National Laboratories – INL, ANL, and LANL.

21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS↗

A Parallel Graph Environment for Real-World Data Analytics Workflows

Economic competitiveness and national security depend increasingly on the insightful analysis of large data sets. The diversity of real-world data sources and analytic workflows impose challenging hardware and software requirements for parallel graph platforms. The irregular nature of graph methods is not supported well by the deep memory hierarchies of conventional distributed systems, requiring new processor and runtime system designs to tolerate memory and synchronization latencies. Moreover, the efficiency of relational table operations and matrix computations are not attainable when data is stored in common graph data structures. In this paper, we present HAGGLE, a high-performance, scalable data analytics platform. The platform’s hybrid data model supports a variety of distributed, thread-safe data structures, parallel programming constructs, and persistent and streaming data. An abstract runtime layer enables us to map the stack to conventional, distributed computer systems with accelerators. The runtime uses multithreading, active messages, and data aggregation to hide memory and synchronization latencies on large-scale systems.

Graph Analytics↗

Defining the S-Glutathionylation Proteome by Biochemical and Mass Spectrometric Approaches

Protein S-glutathionylation (SSG) is a reversible post-translational modification (PTM) featuring the conjugation of glutathione to a protein cysteine thiol. SSG can alter protein structure, activity, subcellular localization, and interaction with small molecules and other proteins. Thus, it plays a critical role in redox signaling and regulation in various physiological activities and pathological events. In this review, we summarize current biochemical and analytical approaches for characterizing SSG at both the proteome level and at individual protein levels. To illustrate the mechanism underlying SSG-mediated redox regulation, we highlight recent examples of functional and structural consequences of SSG modifications. Finally, we discuss the analytical challenges in characterizing SSG and the thiol PTM landscape, future directions for understanding of the role of SSG in redox signaling and regulation and its interplay with other PTMs, and the potential role of computational approaches to accelerate functional discovery.

59 BASIC BIOLOGICAL SCIENCES↗

Enabling Parallel Performance and Portability of Solid Mechanics Simulations Across CPU and GPU Architectures

Efficiently simulating solid mechanics is vital across various engineering applications. As constitutive models grow more complex and simulations scale up in size, harnessing the capabilities of modern computer architectures has become essential for achieving timely results. This paper presents advancements in running parallel simulations of solid mechanics on multi-core CPUs and GPUs using a single-code implementation. This portability is made possible by the C++ matrix and array (MATAR) library, which interfaces with the C++ Kokkos library, enabling the selection of fine-grained parallelism backends (e.g., CUDA, HIP, OpenMP, pthreads, etc.) at compile time. MATAR simplifies the transition from Fortran to C++ and Kokkos, making it easier to modernize legacy solid mechanics codes. We applied this approach to modernize a suite of constitutive models and to demonstrate substantial performance improvements across different computer architectures. This paper includes comparative performance studies using multi-core CPUs along with AMD and NVIDIA GPUs. Results are presented using a hypoelastic–plastic model, a crystal plasticity model, and the viscoplastic self-consistent generalized material model (VPSC-GMM). The results underscore the potential of using the MATAR library and modern computer architectures to accelerate solid mechanics simulations.

Morgan, Nathaniel (ORCID:0000000276118449)↗

Beam efflux measurements

Measurements have been made of the high energy thrust ions, (Group I), high angle/high energy ions (Group II), and high angle/low energy ions (Group IV) of a mercury electron bombardment thruster in the angular divergence range from 0 deg to greater than 90 deg. The measurements have been made as a function of thrust ion current, propellant utilization efficiency, bombardment discharge voltage, screen and accelerator grid potential (accel-decel ratio) and neutralizer keeper potential. The shape of the Group IV (charge exchange) ion plume has remained essentially fixed within the range of variation of the engine operation parameters. The magnitude of the charge exchange ion flux scales with thrust ion current, for good propellant utilization conditions. For fixed thrust ion current, charge exchange ion flux increases for diminishing propellant utilization efficiency. Facility effects influence experimental accuracies within the range of propellant utilization efficiency used in the experiments. The flux of high angle/high energy Group II ions is significantly diminished by the use of minimum decel voltages on the accelerator grid. A computer model of charge exchange ion production and motion has been developed. The program allows computation of charge exchange ion volume production rate, total production rate, and charge exchange ion trajectories for "genuine" and "facilities effects" particles. In the computed flux deposition patterns, the Group I and Group IV ion plumes exhibit a counter motion.

Komatsu, G. K.↗

Algorithms for computing the geopotential using a simple density layer

Several algorithms have been developed for computing the potential and attraction of a simple density layer. These are numerical cubature, Taylor series, and a mixed analytic and numerical integration using a singularity-matching technique. A computer program has been written to combine these techniques for computing the disturbing acceleration on an artificial earth satellite. A total of 1640 equal-area, constant surface density blocks on an oblate spheroid are used. The singularity-matching algorithm is used in the subsatellite region, Taylor series in the surrounding zone, and numerical cubature on the rest of the earth.

Morrison, F.↗

Simplified and refined structural modeling for economical flutter analysis and design

A coordinated use of two finite-element models of different levels of refinement is presented to reduce the computer cost of the repetitive flutter analysis commonly encountered in structural resizing to meet flutter requirements. One model, termed a refined model (RM), represents a high degree of detail needed for strength-sizing and flutter analysis of an airframe. The other model, called a simplified model (SM), has a relatively much smaller number of elements and degrees-of-freedom. A systematic method of deriving an SM from a given RM is described. The method consists of judgmental and numerical operations to make the stiffness and mass of the SM elements equivalent to the corresponding substructures of RM. The structural data are automatically transferred between the two models. The bulk of analysis is performed on the SM with periodical verifications carried out by analysis of the RM. In a numerical example of a supersonic cruise aircraft with an arrow wing, this approach permitted substantial savings in computer costs and acceleration of the job turn-around.

Ricketts, R. H.↗

Shuttle /STS-1/ entry trajectory reconstruction

The methods and data used to reconstruct the Shuttle entry trajectory aerodynamics from 180 km to landing are reviewed. Onboard accelerometer and gyro measurements are used with a weighted least squares filter along with ground tracking data to provide the evolving spacecraft state. The software development is described, noting that computations include linear acceleration and angular rate data, aerodynamic coefficient, and inertial measurement units, sampled at 170 Hz. Ground tracking data was gathered on S- and C-bands; Guam data is considered critical as it was taken during the blackout at the beginning of entry. The generation of a best estimate trajectory, combining flight and ground data, yields inplane parameters of altitude, flight path angle, and velocity for comparison with navigational instrumentation. Predicted lift/drag ratios were found to be accurate from Mach 1-25, and up to 23% low below Mach 1.

Compton, H. R.↗

Implicit upwind methods for the compressible Navier-Stokes equations

A class of implicit upwind-differencing methods for the compressible Navier-Stokes equations is described and applied. The methods are based on the use of local eigenvalues or wave speeds to control spatial differencing of inviscid terms and are aimed at increasing the level of accuracy and stability achievable in computation. Techniques for accelerating the rate of convergence to a steady-state solution are also used. Applications to inviscid and viscous transonic flows are discussed and compared with other methods and experimental measurements. It is shown that accurate and efficient transonic airfoil calculations can be made on the Cray-1 coomputer in less than 2 min.

Coakley, T. J.↗

Independent pixel and Monte Carlo estimates of stratocumulus albedo

Monte Carlo radiative transfer methods are employed here to estimate the plane-parallel albedo bias for marine stratocumulus clouds. This is the bias in estimates of the mesoscale-average albedo, which arises from the assumption that cloud liquid water is uniformly distributed. The authors compare such estimates with those based on a more realistic distribution generated from a fractal model of marine stratocumulus clouds belonging to the class of 'bounded cascade' models. In this model the cloud top and base are fixed, so that all variations in cloud shape are ignored. The model generates random variations in liquid water along a single horizontal direction, forming fractal cloud streets while conserving the total liquid water in the cloud field. The model reproduces the mean, variance, and skewness of the vertically integrated cloud liquid water, as well as its observed wavenumber spectrum, which is approximately a power law. The Monte Carlo method keeps track of the three-dimensional paths solar photons take through the cloud field, using a vectorized implementation of a direct technique. The simplifications in the cloud field studied here allow the computations to be accelerated. The Monte Carlo results are compared to those of the independent pixel approximation, which neglects net horizontal photon transport. Differences between the Monte Carlo and independent pixel estimates of the mesoscale-average albedo are on the order of 1% for conservative scattering, while the plane-parallel bias itself is an order of magnitude larger. As cloud absorption increases, the independent pixel approximation agrees even more closely with the Monte Carlo estimates. This result holds for a wide range of sun angles and aspect ratios. Thus, horizontal photon transport can be safely neglected in estimates of the area-average flux for such cloud models. This result relies on the rapid falloff of the wavenumber spectrum of stratocumulus, which ensures that the smaller-scale variability, where the radiative transfer is more three-dimensional, contributes less to the plane-parallel albedo bias than the larger scales, which are more variable. The lack of significant three-dimensional effects also relies on the assumption of a relatively simple geometry. Even with these assumptions, the independent pixel approximation is accurate only for fluxes averaged over large horizontal areas, many photon mean free paths in diameter, and not for local radiance values, which depend strongly on the interaction between neighboring cloud elements.

Cahalan, Robert F.↗

Predicting Arrival Of Protons Emitted In Solar Flares

Visual Utility for Localization of Corona Accelerated Nuclei (VULCAN) computer program provides both advance warnings and insight for post-event analyses of effects of solar flares. Using measurements of peak fluxes, times of detection, flare location, solar wind velocities, and x-ray emissions from Sun, as electronically sent by NOAA (National Oceanographic and Atmospheric Administration), VULCAN predicts resulting intensities of proton fluxes at various user-chosen points (spacecraft or planets) of solar system. Also predicts times of onset of fluxes of protons and peak values of fluxes.

Spagnuolo, John N., Jr.↗

Effects of Convection During the Photodeposition of Polydiacetylene Thin Films

In this work, we describe a preliminary investigation of buoyancy-driven heat transfer during the growth of thin films from solution following exposure to ultraviolet (UV) light. Irradiation of the growth cell occurs at various directions relative to gravitational acceleration. Through numerical computations, the steady-state flow and temperature profiles are simulated during the course of light exposure. Light-induced polymerization accompanies a heat transfer process through a fairly complicated recirculating flow pattern. A scaling analysis shows that buoyancy-driven velocities only reduce by a factor of 10 for gravity levels as low as 10(exp -2)g(sub 0). Paley et al. observe what appears to be gravitationally sensitive particle development and inclusion in thin films using a photodeposition process. From this study it is clear that production of homogeneous thin films would have to occur in the environment of a complicated flow pattern of recirculation with a nonuniform temperature distribution. Indeed, even when irradiation occurs from the top of the cell, the most stable stratified cell orientation, defects remain in our films due to the persistence of buoyancy-driven convection. To achieve homogeneity, minimal scattering centers, and possible molecular order, photodeposition of polymer films by UV light exposure must proceed in a reduced-convection environment. Fluid mechanics simulations are useful for establishing gravitational sensitivity to this recently discovered process (patent # 5,451,433) for preparing thin films having quite promising nonlinear optical characteristics.

Frazier, D. O.↗

Effects of Convection during the Photodeposition of Polydiacetylene Thin Films

In this work, we describe a preliminary investigation of buoyancy-driven heat transfer during the growth of thin films from solution following exposure to ultraviolet (UV) light. Irradiation of the growth cell occurs at various directions relative to gravitational acceleration. Through numerical computations, the steady-state flow and temperature profiles are simulated during the course of light exposure. Light-induced polymerization accompanies a heat transfer process through a fairly complicated recirculating flow pattern. A scaling analysis shows that buoyancy-driven velocities only reduce by a factor of 10 for gravity levels as low as 10(exp -2) g(sub 0). Paley et al. observe what appears to be gravitationally sensitive particle development and inclusion in thin films using a photodeposition process. From this study, it is clear that production of homogeneous thin films would have to occur in the environment of a complicated flow pattern of recirculation with a nonuniform temperature distribution. Indeed, even when irradiation occurs from the top of the cell, the most stable stratified cell orientation, defects remain in our films due to the persistence of buoyancy-driven convection. To achieve homogeneity, minimal scattering centers, and possible molecular order, photodeposition of polymer films by UV light exposure must proceed in a reduced-convection environment. Fluid mechanics simulations are useful for establishing gravitational sensitivity to this recently discovered process (patent # 5,451,433) for preparing thin films having quite promising nonlinear optical characteristics.

Frazier, D. O.↗

Cosmic Rays and Their Radiative Processes in Numerical Cosmology

A cosmological hydrodynamic code is described, which includes a routine to compute cosmic ray acceleration and transport in a simplified way. The routine was designed to follow explicitly diffusive, acceleration at shocks, and second-order Fermi acceleration and adiabatic loss in smooth flows. Synchrotron cooling of the electron population can also be followed. The updated code is intended to be used to study the properties of nonthermal synchrotron emission and inverse Compton scattering from electron cosmic rays in clusters of galaxies, in addition to the properties of thermal bremsstrahlung emission from hot gas. The results of a test simulation using a grid of 128 (exp 3) cells are presented, where cosmic rays and magnetic field have been treated passively and synchrotron cooling of cosmic ray electrons has not been included.

Ryu, Dongsu↗

Micron-Accurate Laser Fresnel-Diffraction Ranging System

Two versions of an optoelectronic system undergoing development are depicted. The system is expected to be capable of measuring a distance between 2 and 10 m with an error of no more than 1 micrometer. The system would be designed to exploit Fresnel diffraction of a laser beam. In particular, it would be designed to take advantage of the fact that a Fresnel diffraction pattern is ultrasensitive to distance. The two versions would differ in the following respects: In version 1, the focus of the telescope would be in the Fresnel region, and the telescope would have a small depth of focus. As a consequence, the Fresnel pattern would be imaged directly onto the photodetector array; in version 2, a multielement lens module would displace the Fresnel region from the vicinity of the pinhole to the vicinity of the optical receiver. As the distance to be measured varied, the location of the receiver relative to the displaced Fresnel-diffraction region would vary, thereby causing the Fresnel diffraction pattern on the focal plane to vary. The multielement lens module would also correct for aberrations. The processing of the digitized Fresnel diffraction pattern in the computer might be accelerated by using only parts of the pattern or even only one small part - the central pixel. As the distance from the pinhole increased, the central pixel would rapidly cycle between maximum and minimum light intensity. This in itself would not be sufficient to uniquely determine the distance. However, by varying the size of the pinhole or the wavelength of the laser, one could obtain a second cycle of variation of intensity that, in conjunction with the first cycle, could enable a unique determination of distance. Alternatively, for a single wavelength and a single pinhole size, it should suffice to consider the data from only two different key pixels in the Fresnel pattern.

Lehner, David↗