Search NASA⌕ Search

SEARCH · Search NASA

Results for “parallel simulation”

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 1,657 records · Page 92

Formation of electron kappa distributions due to interactions with parallel propagating whistler waves

In space plasmas, charged particles are frequently observed to possess a high-energy tail, which is often modeled by a kappa-type distribution function. In this work, the formation of the electron kappa distribution in generation of parallel propagating whistler waves is investigated using fully nonlinear particle-in-cell (PIC) simulations. A previous research concluded that the bi-Maxwellian character of electron distributions is preserved in PIC simulations. We now demonstrate that for interactions between electrons and parallel propagating whistler waves, a non-Maxwellian high-energy tail can be formed, and a kappa distribution can be used to fit the electron distribution in time-asymptotic limit. The κ-parameter is found to decrease with increasing initial temperature anisotropy or decreasing ratio of electron plasma frequency to cyclotron frequency. The results might be helpful to understanding the origin of electron kappa distributions observed in space plasmas.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Study of the effective point of measurement for ion chambers in electron beams by Monte Carlo simulation

In current dosimetry protocols for electron beams, for plane-parallel chambers, the effective point of measurement is at the front face of the cavity, and, for cylindrical chambers, it is at a point shifted 0.5r upstream from the cavity center. In this study, Monte Carlo simulations are employed to study the issue of effective point of measurement for both plane-parallel chambers and cylindrical thimble chambers in electron beams. It is found that there are two ways of determining the position of the effective point of measurement: One is to match the calculated depth-ionization curve obtained from a modeled chamber to a calculated depth-dose curve; the other is to match the electron fluence spectrum in the chamber cavity to that in the phantom. For plane-parallel chambers, the effective point of measurement determined by the first method is generally not at the front face of the chamber cavity, which is obtained by the second method, but shifted downstream toward the cavity center by an amount that could be larger than one-half a millimeter. This should not be ignored when measuring depth-dose curves in electron beams. For cylindrical chambers, these two methods also give different positions of the effective point of measurement: The first gives a shift of 0.5r, which is in agreement with measurements for high-energy beams and is the same as the value currently used in major dosimetry protocols; the latter gives a shift of 0.8r, which is closer to the value predicted by a theoretical calculation assuming no-scatter conditions. The results also show that the shift of 0.8r is more appropriate if the cylindrical chamber is to be considered as a Spencer-Attix cavity. In electron beams, since the water/air stopping-power ratio changes with depth in a water phantom, the difference of the two shifts (0.3r) will lead to an incorrect evaluation of the water/air stopping-power ratio at the point of measurement, thus resulting in a systematic error in determining the absorbed dose by cylindrical chambers. It is suggested that a shift of 0.8r be used for electron beam calibrations with cylindrical chambers and a shift of 0.4r-0.5r be used for depth-dose measurements.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

toolkit computational mesh conceptual model.

The Sierra Toolkit computational mesh is a software library intended to support massively parallel multi-physics computations on dynamically changing unstructured meshes. This domain of intended use is inherently complex due to distributed memory parallelism, parallel scalability, heterogeneity of physics, heterogeneous discretization of an unstructured mesh, and runtime adaptation of the mesh. Management of this inherent complexity begins with a conceptual analysis and modeling of this domain of intended use; i.e., development of a domain model. The Sierra Toolkit computational mesh software library is designed and implemented based upon this domain model. Software developers using, maintaining, or extending the Sierra Toolkit computational mesh library must be familiar with the concepts/domain model presented in this report.

97 MATHEMATICS AND COMPUTING↗

NASA Tech Briefs, February 2014

Topics include: JWST Integrated Simulation and Test (JIST) Core; Software for Non-Contact Measurement of an Individual's Heart Rate Using a Common Camera; Rapid Infrared Pixel Grating Response Testbed; Temperature Measurement and Stabilization in a Birefringent Whispering Gallery Resonator; JWST IV and V Simulation and Test (JIST) Solid State Recorder (SSR) Simulator; Development of a Precision Thermal Doubler for Deep Space; Improving Friction Stir Welds Using Laser Peening; Methodology of Evaluating Margins of Safety in Critical Brazed Joints; Interactive Inventory Monitoring; Sensor for Spatial Detection of Single-Event Effects in Semiconductor-Based Electronics; Reworked CCGA-624 Interconnect Package Reliability for Extreme Thermal Environments; Current-Controlled Output Driver for Directly Coupled Loads; Bulk Metallic Glasses and Matrix Composites as Spacecraft Shielding; Touch Temperature Coating for Electrical Equipment on Spacecraft; Li-Ion Electrolytes Containing Flame-Retardant Additives; Autonomous Robotic Manipulation (ARM); CARVE Log; Platform Perspective Toolkit; Convex Hull-Based Plume and Anomaly Detection; Pre-Filtration of GOSAT Data Using Only Level 1 Data and an Intelligent Filter to Remove Low Clouds; Affordability Comparison Tool - ACT; "Ascent - Commemorating Shuttle" for iPad; Cassini Mission App; Light-Weight Workflow Engine: A Server for Executing Generic Workflows; Model for System Engineering of the CheMin Instrument; Timeline Central Concepts; Parallel Particle Filter Toolkit; Particle Filter Simulation and Analysis Enabling Non-Traditional Navigation; Quasi-Terminator Orbits for Mapping Small Primitive Bodies; The Subgrid-Scale Scalar Variance Under Supercritical Pressure Conditions; Sliding Gait for ATHLETE Mobility; and Automated Generation of Adaptive Filter Using a Genetic Algorithm and Cyclic Rule Reduction.

Source record↗

Large-Scale Multi-Dimensional Document Clustering on GPU Clusters

Document clustering plays an important role in data mining systems. Recently, a flocking-based document clustering algorithm has been proposed to solve the problem through simulation resembling the flocking behavior of birds in nature. This method is superior to other clustering algorithms, including k-means, in the sense that the outcome is not sensitive to the initial state. One limitation of this approach is that the algorithmic complexity is inherently quadratic in the number of documents. As a result, execution time becomes a bottleneck with large number of documents. In this paper, we assess the benefits of exploiting the computational power of Beowulf-like clusters equipped with contemporary Graphics Processing Units (GPUs) as a means to significantly reduce the runtime of flocking-based document clustering. Our framework scales up to over one million documents processed simultaneously in a sixteennode GPU cluster. Results are also compared to a four-node cluster with higher-end GPUs. On these clusters, we observe 30X-50X speedups, which demonstrates the potential of GPU clusters to efficiently solve massive data mining problems. Such speedups combined with the scalability potential and accelerator-based parallelization are unique in the domain of document-based data mining, to the best of our knowledge.

99 GENERAL AND MISCELLANEOUS↗

Time-dependent simulation of cosmic-ray shocks, including Alfven transport

Time evolution of plane, cosmic-ray modified shocks has been simulated numerically for the case with parallel magnetic fields. Computations were done in a 'three-fluid' dynamical model incorporating cosmic-ray and Alfven-wave energy transport equations. Nonlinear feedback from the cosmic rays and Alfven waves is included in the equation of motion for the underlying plasma, as is the finite propagation speed and energy dissipation of the Alfven waves. Exploratory results confirm earlier, steady state analyses that found these Alfven transport effects to be potentially important when the upstream Alfven speed and gas sound speeds are comparable. As noted earlier, Alfven transport effects tend to reduce the transfer of energy through a shock from gas to energetic particles. These studies show as well that the timescale for modification of the shock is altered in nonlinear ways. It is clear, however, that the consequences of Alfven transport are strongly model dependent and that both advection of cosmic rays by the waves and dissipation of wave energy in the plasma will be important to model correctly when quantitative results are needed. Comparison is made between simulations based on a constant diffusion coefficient and more realistic diffusion models allowing the diffusion coefficient to vary in response to changes in Alfven wave intensity. No really substantive differences were found between them.

Jones, T. W.↗

Time dependent simulation of cosmic-ray shocks including Alfven transport

Time evolution of plane, cosmic-ray modified shocks was simulated numerically for the case with parallel magnetic fields. Computations were done in a 'three-fluid' dynamical model incorporating cosmic-ray and Alfven wave energy transport equations. Nonlinear feedback from the cosmic-rays and Alfven waves is included in the equation of motion for the underlying plasma, as is the finite propagation speed and energy dissipation of the Alfven waves. Exploratory results confirm earlier, steady state analyses that found these Alfven transport effects to be potentially important when the upstream Alfven speed and gas sound speeds are comparable. As noted earlier Alfven transport effects tend to reduce the transfer of energy through a shock from gas to energetic particles. These studies show as well that the time scale for modification of the shock is altered in nonlinear ways. It is clear, however, that the consequences of Alfven transport are strongly model dependent and that both advection of cosmic-rays by the waves and dissipation of wave energy in the plasma will be important to model correctly when quantitative results are needed. Comparison is made between simulations based on a constant diffusion coefficient and more realistic diffusion models allowing the diffusion coefficient to vary in response to changes in Alfven wave intensity. No really substantive differences were found between them.

Jones, T. W.↗

Exploiting New Features of COMSOL Version 4 on Conjugate Heat Transfer Problems

Users of COMSOL Multiphysics at version 3.5a and earlier have enjoyedmany features that have provided not only a good user experience atthe GUI interface, but also the capability to solve many classes ofproblems in a consistent manner with the physics being simulated. Withthe new release version 4.0 and later (4+) of COMSOL, the user isprovided a dramatic new interface from which to interact, and many newfeatures ``under the hood'' for solving problems more efficiently and with evengreater accuracy and consistency than before. This paper will exploreseveral of these new version 4+ features for the conjugate heattransfer class of problems. Our environment is challenging in that wedemand high-quality solutions for nuclear-reactor systems and themodels tend to become large and difficult to solve. Areasinvestigated include turbulence modeling, distributed parallelprocessing, solver scaling, and opengl graphics issues in a Linuxcomputing environment.

99 GENERAL AND MISCELLANEOUS↗

Numerical Simulation of Protoplanetary Vortices

The fluid dynamics within a protoplanetary disk has been attracting the attention of many researchers for a few decades. Previous works include, to list only a few among many others, the well-known prescription of Shakura & Sunyaev, the convective and instability study of Stone & Balbus and Hawley et al., the Rossby wave approach of Lovelace et al., as well as a recent work by Klahr & Bodenheimer, which attempted to identify turbulent flow within the disk. The disk is commonly understood to be a thin gas disk rotating around a central star with differential rotation (the Keplerian velocity), and the central quest remains as how the flow behavior deviates (albeit by a small amount) from a strong balance established between gravitational and centrifugal forces, transfers mass and momentum inward, and eventually forms planetesimals and planets. In earlier works we have briefly described the possible physical processes involved in the disk; we have proposed the existence of long-lasting, coherent vortices as an efficient agent for mass and momentum transport. In particular, Barranco et al. provided a general mathematical framework that is suitable for the asymptotic regime of the disk; Barranco & Marcus (2000) addressed a proposed vortex-dust interaction mechanism which might lead to planetesimal formation; and Lin et al. (2002), as inspired by general geophysical vortex dynamics, proposed basic mechanisms by which vortices can transport mass and angular momentum. The current work follows up on our previous effort. We shall focus on the detailed numerical implementation of our problem. We have developed a parallel, pseudo-spectral code to simulate the full three-dimensional vortex dynamics in a stably-stratified, differentially rotating frame, which represents the environment of the disk. Our simulation is validated with full diagnostics and comparisons, and we present our results on a family of three-dimensional, coherent equilibrium vortices.

Lin, H.↗

Combining machine-learned and empirical force fields with the parareal algorithm: application to the diffusion of atomistic defects

We numerically investigate an adaptive version of the parareal algorithm in the context of molecular dynamics. This adaptive variant has been originally introduced in [1]. We focus here on test cases of physical interest where the dynamics of the system is modelled by the Langevin equation and is simulated using the molecular dynamics software LAMMPS. In this work, the parareal algorithm uses a family of machine-learning spectral neighbor analysis potentials (SNAP) as fine, reference, potentials and embedded-atom method potentials (EAM) as coarse potentials. We consider a self-interstitial atom in a tungsten lattice and compute the average residence time of the system in metastable states. Our numerical results demonstrate significant computational gains using the adaptive parareal algorithm in comparison to a sequential integration of the Langevin dynamics. We also identify a large regime of numerical parameters for which statistical accuracy is reached without being a consequence of trajectorial accuracy.

36 MATERIALS SCIENCE↗

TEST FOR WAVEVECTOR ANISOTROPIES IN PLASMA TURBULENCE CASCADES

The frequency and wavevector matching conditions in nonlinear three-wave coupling are used to test whether the forward cascade of plasma turbulence may lead to wavevector anisotropies in a homogeneous, collisionless, magnetized plasma. Linear kinetic theory at {beta}{sub p} = 0.01, 0.10, and 1.0 is used to determine the frequency-wavenumber dispersion of three normal modes: long-wavelength Alfven-cyclotron waves, long-wavelength magnetosonic waves, and intermediate-wavelength magnetosonic-whistler waves. Using linear dispersion in the nonlinear matching conditions, the test predicts with one exception that forward cascades are favored by fluctuations propagating nearly perpendicular to the background magnetic field B{sub o}. This is consistent with the typical development of wavevector anisotropies with k >> k{sub Parallel-To} (subscripts refer to directions perpendicular and parallel to B{sub o}, respectively) in computer simulations of the forward cascade of various types of plasma turbulence. The exception is that, at {beta}{sub p} = 1.0, the test predicts that the cascade of long-wavelength magnetosonic waves should be favored by modes at k {approx} k{sub Parallel-To }.

79 ASTRONOMY AND ASTROPHYSICS↗

Rendezvous algorithms for large-scale modeling and simulation

Rendezvous algorithms encode a communication pattern that is useful when processors sending data do not know who the receiving processors should be, or vice versa. The idea is to define an intermediate decomposition where datums from different sending processors can ”rendezvous” to perform a computation, in a manner that both the senders and eventual receivers of the results can identify the appropriate rendezvous processor. Though they were originally designed for interpolating between overlaid grids with independent parallel decompositions (Plimpton et al., 2004), we have recently found rendezvous algorithms useful for a variety of operations in particle- or grid-based simulation codes when running large problems on large numbers of processors. In particular, we show they can perform well when a load-balanced intermediate decomposition is randomized and not spatial, requiring all-to-all communication to move data between processors. In this case rendezvous algorithms leverage the large bisection communication bandwidths which parallel machines provide. We describe how rendezvous algorithms work in a scientific computing context and give specific examples for molecular dynamics and Direct Simulation Monte Carlo codes which result in dramatic performance improvements versus simpler algorithms which do not scale as well. We explain how a generic rendezvous algorithm can be implemented, and also point out similarities with the MapReduce paradigm popularized by Google and Hadoop.

97 MATHEMATICS AND COMPUTING↗

Modeling of a negative ion source. III. Two-dimensional structure of the extraction region

The self-consistent production and transport of H{sup -} in the extraction region of a hybrid negative ion source is modeled by means of a two-dimensional particle-in-cell/Monte Carlo simulation. The normal coordinate and one parallel coordinate with respect to the plasma grid are considered to analyze the transport of negative ions. Results show that, in order to establish space charge compensation, the extraction of surface-produced negative ions is limited by the flux of positive ions directed toward the plasma grid surface. An electrostatic barrier appears just in front of the wall, reflecting the majority of surface-produced H{sup -} and reducing by this their extraction probability to only 8.5%. Results reproduce the experimentally observed influence of the plasma grid bias voltage on the extraction identifying as a key element the presence of a saddle point in the electric potential distribution.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Compressing unstructured mesh data from simulations using machine learning

The amount of data output from a computer simulation has grown to terabytes and petabytes as increasingly complex simulations are being run on massively parallel systems. As we approach exaflop computing in the next decade, it is expected that the I/O subsystem will not be able to write out these large volumes of data. In this paper, we explore the use of machine learning to compress the data before it is written out. Despite the computational constraints that limit us to using very simple learning algorithms, our results show that machine learning is a viable option for compressing unstructured data. Furthermore, we demonstrate that by simply using a better sampling algorithm to generate the training set, we can obtain more accurate results compared to random sampling, but at no extra cost. Further, by carefully selecting and incorporating points with high prediction error, we can improve reconstruction accuracy without sacrificing the compression rate.

97 MATHEMATICS AND COMPUTING↗

Parallel sorting algorithm classification: is manual instrumentation necessary?

Understanding parallel algorithms is crucial for accelerating scientific simulations on complex, distributed memory, high-performance computers. Modern algorithm classification approaches learn semantics directly from source code to differentiate between algorithms, however, accessing source code is not always possible. We can learn about parallel algorithms from observing their performance, as programs running the same algorithms and using the same hardware should exhibit similar performance characteristics. We present an approach to learn algorithm classes from parallel performance data directly in order to classify algorithms without access to the source code. We extend previous work to enable classifying parallel sorting algorithms using automatic instrumentation instead of requiring manual region annotations in the source code. In this work, we design and demonstrate a study for classification of parallel sorting algorithms using parallel performance data collected from automatic instrumentation, and evaluate the performance of our new methodology on classification. We leverage Caliper to collect the performance data, Thicket for our exploratory data analysis (EDA), and PyTorch and Scikit-learn to evaluate the effectiveness of random forests, support vector machines (SVMs), decision trees, neural networks, and logistic regressions on parallel performance data. Additionally, we study noise in parallel performance data, whether the removal of noise and pre-processing of the data is necessary to accurately classify parallel sorting algorithms, and determine the effectiveness of features created from performance data. In conclusion, we demonstrate classification accuracy for these five different models of up to 97.7% across four different parallel algorithm classes.

Algorithm Classification↗

Advances in computational design and analysis of airbreathing propulsion systems

The development of commercial and military aircraft depends, to a large extent, on engine manufacturers being able to achieve significant increases in propulsion capability through improved component aerodynamics, materials, and structures. The recent history of propulsion has been marked by efforts to develop computational techniques that can speed up the propulsion design process and produce superior designs. The availability of powerful supercomputers, such as the NASA Numerical Aerodynamic Simulator, and the potential for even higher performance offered by parallel computer architectures, have opened the door to the use of multi-dimensional simulations to study complex physical phenomena in propulsion systems that have previously defied analysis or experimental observation. An overview of several NASA Lewis research efforts is provided that are contributing toward the long-range goal of a numerical test-cell for the integrated, multidisciplinary design, analysis, and optimization of propulsion systems. Specific examples in Internal Computational Fluid Mechanics, Computational Structural Mechanics, Computational Materials Science, and High Performance Computing are cited and described in terms of current capabilities, technical challenges, and future research directions.

Klineberg, John M.↗

Advances in computational design and analysis of airbreathing propulsion systems

The development of commercial and military aircraft depends, to a large extent, on engine manufacturers being able to achieve significant increases in propulsion capability through improved component aerodynamics, materials, and structures. The recent history of propulsion has been marked by efforts to develop computational techniques that can speed up the propulsion design process and produce superior designs. The availability of powerful supercomputers, such as the NASA Numerical Aerodynamic Simulator, and the potential for even higher performance offered by parallel computer architectures, have opened the door to the use of multi-dimensional simulations to study complex physical phenomena in propulsion systems that have previously defied analysis or experimental observation. An overview of several NASA Lewis research efforts is provided that are contributing toward the long-range goal of a numerical test-cell for the integrated, multidisciplinary design, analysis, and optimization of propulsion systems. Specific examples in Internal Computational Fluid Mechanics, Computational Structural Mechanics, Computational Materials Science, and High Performance Computing are cited and described in terms of current capabilities, technical challenges, and future research directions.

Klineberg, John M.↗

Nonlocal, diamagnetic electromagnetic effects in magnetically insulated transmission lines

We identify the time-dependent physics responsible for the critical reduction of current losses in magnetically insulated transmission lines (MITLs) due to uninsulated space charge-limited currents of electrons emitted by field stress. A drive current of sufficiently short pulse length introduces a strong enough time dependence that steady-state results alone become inadequate for the complete understanding of current losses. The time-dependent physics can be described as a nonlocal, diamagnetic electromagnetic response of space charge limited currents. As the pulse length is increased or equivalently, the MITL length reduced, these time-dependent effects diminish and current losses converge to those predicted by the well-known Child–Langmuir law in the external (vacuum) fields. We present a simple one-dimensional (1D) model that encapsulates the essence of this physics. We find excellent agreement with 2D particle-in-cell simulations for two MITL geometries, Cartesian parallel plate and azimuthally symmetric straight coaxial. Based on the 1D model, we explore various scaling dependencies of MITL losses with relevant parameters, e.g., peak current, pulse length, geometrical dimensions, etc. We propose an improved physics model of magnetic insulation in the form of a Hull curve, which could also help improve predictions of current losses by common circuit element codes, such as BERTHA. Finally, we describe how to calculate the temperature rise due to electron impact within the 1D model.

Computer simulation↗