Search NASA⌕ Search

SEARCH · Search NASA

Results for “distributed parallelization”

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 1,117 records · Page 62

Parallel Optical Flow Processing of 4D Cardiac CT Data on Multicore Clusters, In: 2014 IEEE 17th International Conference on Computational Science and Engineering

Optical flow is the distribution of apparent velocities of movement of brightness patterns in a sequence of images. For large 3D image sequences, optical flow applications are time consuming and memory-bound. To cope with these problems, in this paper, we present parallel optical flow processing of 4D cardiac CT data on multicore cluster systems to significantly shorten the time for computing velocity fields of the heart in order to aid cardiologists in diagnosing heart disease such as myocardial infarction and cardiac dysrhythmia in time. First, we modify and extend two traditional 2D optical flow methods Horn-Schunck and Lucas-Kanade to three-dimensional cases to process the 4D cardiac CT data. Second, we extend Mat lab MPI to support parallel computing with Mat lab and Octave on these cluster systems. Then we develop the parallel Mat lab/Octave optical flow applications for the 4D cardiac CT data in detail. Our experimental results show that these parallel optical flow applications have good scalability with close to linear speedup, and are able to shorten the image processing time significantly from more than 5 hours on 4 cores to 1.5 minutes on 1024 cores.

Engineering↗

Small tensor product distributed active space (STP-DAS) framework for relativistic and non-relativistic multiconfiguration calculations: Scaling from 10 9 on a laptop to 10 12 determinants on a supercomputer

Despite the power and flexibility of configuration interaction (CI) based methods in computational chemistry, their broader application is limited by an exponential increase in both computational and storage requirements, particularly due to the substantial memory needed for excitation lists that are crucial for scalable parallel computing. Here, the objective of this work is to develop a new CI framework, namely, the small tensor product distributed active space (STP-DAS) framework, aimed at drastically reducing memory demands for extensive CI calculations on individual workstations or laptops, while simultaneously enhancing scalability for extensive parallel computing. Moreover, the STP-DAS framework can support various CI-based techniques, such as complete active space (CAS), restricted active space, generalized active space, multireference CI, and multireference perturbation theory, applicable to both relativistic (two- and four-component) and non-relativistic theories, thus extending the utility of CI methods in computational research. We conducted benchmark studies on a supercomputer to evaluate the storage needs, parallel scalability, and communication downtime using a realistic exact-two-component CASCI (X2C-CASCI) approach, covering a range of determinants from 10 9 to 10 12 . Additionally, we performed large X2C-CASCI calculations on a single laptop and examined how the STP-DAS partitioning affects performance.

Complete-active space self-consistent field↗

Compact Graph Representations and Parallel Connectivity Algorithms for Massive Dynamic Network Analysis

Graph-theoretic abstractions are extensively used to analyze massive data sets. Temporal data streams from socioeconomic interactions, social networking web sites, communication traffic, and scientific computing can be intuitively modeled as graphs. We present the first study of novel high-performance combinatorial techniques for analyzing large-scale information networks, encapsulating dynamic interaction data in the order of billions of entities. We present new data structures to represent dynamic interaction networks, and discuss algorithms for processing parallel insertions and deletions of edges in small-world networks. With these new approaches, we achieve an average performance rate of 25 million structural updates per second and a parallel speedup of nearly28 on a 64-way Sun UltraSPARC T2 multicore processor, for insertions and deletions to a small-world network of 33.5 million vertices and 268 million edges. We also design parallel implementations of fundamental dynamic graph kernels related to connectivity and centrality queries. Our implementations are freely distributed as part of the open-source SNAP (Small-world Network Analysis and Partitioning) complex network analysis framework.

97 MATHEMATICS AND COMPUTING↗

Investigation on the similarity law of low-pressure glow discharges based on the light intensity distributions in geometrically similar gaps

Experimental investigation of the light intensity distributions of a low-pressure glow discharge is carried out in several pairs of geometrically similar plane-parallel gaps, of which the aspect ratios and the products of the linear dimension and the gas pressure are the same. The discharge images are captured using a Charge Coupled Device camera, from which the corresponding axial light intensity distributions are presented. Based on the obtained light intensity distributions, the thicknesses of cathode fall layers were identified by measuring the distance between the peak glow position and the cathode boundary. The influence of the discharge current on the light intensity distributions on the geometrically similar gaps is also investigated. It was found that, for discharges in each pair of geometrically similar gaps, the reduced cathode fall thicknesses are observed to be identical when the discharge currents are the same. In conclusion, the similarity relation of the cathode fall thickness is validated for low-pressure glow discharges in gaps for different aspect ratios.

42 ENGINEERING↗

Quality assurance of proton beams using a multilayer ionization chamber system

Purpose: The measurement of percentage depth-dose (PDD) distributions for the quality assurance of clinical proton beams is most commonly performed with a computerized water tank dosimetry system with ionization chamber, commonly referred to as water tank. Although the accuracy and reproducibility of this method is well established, it can be time-consuming if a large number of measurements are required. In this work the authors evaluate the linearity, reproducibility, sensitivity to field size, accuracy, and time-savings of another system: the Zebra, a multilayer ionization chamber system.Methods: The Zebra, consisting of 180 parallel-plate ionization chambers with 2 mm resolution, was used to measure depth-dose distributions. The measurements were performed for scattered and scanned proton pencil beams of multiple energies delivered by the Hitachi PROBEAT synchrotron-based delivery system. For scattered beams, the Zebra-measured depth-dose distributions were compared with those measured with the water tank. The principal descriptors extracted for comparisons were: range, the depth of the distal 90% dose; spread-out Bragg peak (SOBP) length, the region between the proximal 95% and distal 90% dose; and distal-dose fall off (DDF), the region between the distal 80% and 20% dose. For scanned beams, the Zebra-measured ranges were compared with those acquired using a Bragg peak chamber during commissioning.Results: The Zebra demonstrated better than 1% reproducibility and monitor unit linearity. The response of the Zebra was found to be sensitive to radiation field sizes greater than 12.5 × 12.5 cm; hence, the measurements used to determine accuracy were performed using a field size of 10 × 10 cm. For the scattered proton beams, PDD distributions showed 1.5% agreement within the SOBP, and 3.8% outside. Range values agreed within −0.1 ± 0.4 mm, with a maximum deviation of 1.2 mm. SOBP length values agreed within 0 ± 2 mm, with a maximum deviation of 6 mm. DDF values agreed within 0.3 ± 0.1 mm, with a maximum deviation of 0.6 mm. For the scanned proton pencil beams, Zebra and Bragg peak chamber range values demonstrated agreement of 0.0 ± 0.3 mm with a maximum deviation of 1.3 mm. The setup and measurement time for all Zebra measurements was 3 and 20 times less, respectively, compared to the water tank measurements.Conclusions: Our investigation shows that the Zebra can be useful not only for fast but also for accurate measurements of the depth-dose distributions of both scattered and scanned proton beams. The analysis of a large set of measurements shows that the commonly assessed beam quality parameters obtained with the Zebra are within the acceptable variations specified by the manufacturer for our delivery system.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Time-of-flight neutron total scattering with applied electric fields: Ex situ and in situ studies of ferroelectric materials

Characterizing the structural response of functional materials (e.g., piezoelectrics and ferroelectrics) to electric fields is key for the creation of structure-property relationships. Here in this paper, we present a new sample environment and data reduction routines which allow the measurement of time-of-flight neutron total scattering during the in situ or ex situ application of high voltage (<10 kV) to a sample. Instead of utilizing the entire detector space of the diffractometer, only selected regions of detector pixels with scattering at the desired angle to the sample electric field are interrogated, which allows the generation of orientation-dependent reciprocal space patterns and real-space pair distribution functions (PDFs). We demonstrate the method using the relaxor ferroelectric Na 1/2 Bi 1/2 TiO 3 and observe lattice expansion parallel and contraction perpendicular to the electric field for both in situ or ex situ experiments, revealing the irreversible nature of the local scale structural changes to this composition. Additionally, changes in the distributions of nearest neighbor metal-oxygen bond lengths are observed, which have been difficult to observe in previously measured analogous orientation-dependent X-ray PDFs. Lastly, considerations related to sample positioning and background subtraction are discussed, and future research directions are suggested.

36 MATERIALS SCIENCE↗

Accelerating Computation of the Unit Commitment Problem (Presentation)

Production cost models (PCMs) simulate power system operation at hourly (or higher) resolution. While computation times often extend into multiple days, the sequential nature of PCM's makes parallelism difficult. We exploit the persistence of unit commitment decisions to select partition boundaries for simulation horizon decomposition and parallel computation. Partitioned simulations are benchmarked against sequential solutions for optimality and computation time.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Is Monte Carlo embarrassingly parallel?

Monte Carlo is often stated as being embarrassingly parallel. However, running a Monte Carlo calculation, especially a reactor criticality calculation, in parallel using tens of processors shows a serious limitation in speedup and the execution time may even increase beyond a certain number of processors. In this paper the main causes of the loss of efficiency when using many processors are analyzed using a simple Monte Carlo program for criticality. The basic mechanism for parallel execution is MPI. One of the bottlenecks turn out to be the rendez-vous points in the parallel calculation used for synchronization and exchange of data between processors. This happens at least at the end of each cycle for fission source generation in order to collect the full fission source distribution for the next cycle and to estimate the effective multiplication factor, which is not only part of the requested results, but also input to the next cycle for population control. Basic improvements to overcome this limitation are suggested and tested. Also other time losses in the parallel calculation are identified. Moreover, the threading mechanism, which allows the parallel execution of tasks based on shared memory using OpenMP, is analyzed in detail. Recommendations are given to get the maximum efficiency out of a parallel Monte Carlo calculation. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Direct 400 Vdc to 1 Vdc power conversion with input series output parallel connection for data center power supplies

Data centers consume an ever-increasing amount of electricity because of the rapid growth of cloud computing and digital information storage. A high voltage point of load (HV POL) converter is proposed to convert the 400-VDC distribution voltage to 1-VDC within a single stage to increase the power conversion efficiency. A six-phase input series output parallel (ISOP) connected structure is implemented for the HV POL. The symmetrical controlled half bridge current doubler is selected as the converter topology in the ISOP structure. The full load efficiency is improved by 4% points compared with state of the art products. A voltage compensator has been designed in order to meet the strict dynamic voltage regulation requirement. A laboratory prototype has been built, and experimental results have been provided to verify the proposed HV POL with a single power conversion stage can meet the dynamic voltage regulation requirement for an on-board power supply with higher efficiency compared to the conventional architecture.

Cui, Yutian↗

Solar wind magnetic fluctuations and electron non-thermal temperature anisotropy: survey of Wind-SWE-VEIS observations

The solar wind electron velocity distribution function (eVDF) exhibits a variety of non-thermal features that deviate from thermal equilibrium. These deviations from equilibrium provide a local source for electromagnetic fluctuation emissions, including the commonly observed electron whistler-cyclotron and firehose instabilities. We present a systematic analysis of Wind-SWE-VEIS observations of solar wind electron plasma and associated Wind-MFI observed magnetic fluctuations. For the first time using the full solar wind electron distribution and its moments, without separation of the various electron components, we show clear evidence that the temperature anisotropy threshold of the parallel electron cyclotron anisotropic instability bounds solar wind electrons during slow solar wind periods. We also demonstrate that during periods of slow solar wind, collisions—while infrequent—are the dominant mechanism by which solar wind electrons are constrained, leading to isotropization. During fast solar wind periods, magnetic fluctuations and solar wind anisotropies are enhanced above the parallel whistler anisotropic threshold boundary and collisional effects are significantly reduced. Preliminary calculations further show that the oblique electron whistler mirror anisotropic instability bounds both the slow and fast solar wind. Regardless of speed, the solar wind electron thermal anisotropy appears globally bounded by the parallel electron firehose instability for anisotropies T{sub e⊥}/T{sub e||}<1. Our results indicate that collisions, while infrequent, play a necessary role in regulating the solar wind eVDFs. In striking contrast to solar wind ions, solar wind electron plasma, when considered globally as a single eVDF, is only marginally stable with respect to parallel propagating instabilities.

79 ASTRONOMY AND ASTROPHYSICS↗

Ion-driven instabilities in the solar wind: Wind observations of 19 March 2005

Intervals of enhanced magnetic fluctuations have been frequently observed in the solar wind. However, it remains an open question as to whether these waves are generated at the Sun and then transported outward by the solar wind or generated locally in the interplanetary medium. Magnetic field and plasma measurements from the Wind spacecraft under slow solar wind conditions on 19 March 2005 demonstrate seven events of enhanced magnetic fluctuations at spacecraft-frame frequencies somewhat above the proton cyclotron frequency and propagation approximately parallel or antiparallel to the background magnetic field B o . The proton velocity distributions during these events are characterized by two components: a more dense, slower core and a less dense, faster beam. In conclusion, observed plasma parameters are used in a kinetic linear dispersion equation analysis for electromagnetic fluctuations at k x B o = 0; for two events the most unstable mode is the Alfvén-cyclotron instability driven by a proton component temperature anisotropy T ⊥ /T || > 1 (where the subscripts denote directions relative to B o ), and for three events the most unstable mode is the right-hand polarized magnetosonic instability driven primarily by ion component relative flows. Thus, both types of ion anisotropies and both types of instabilities are likely to be local sources of these enhanced fluctuation events in the solar wind.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

A statistical study of EMIC waves observed by Cluster: 2. Associated plasma conditions

This is the second in a pair of papers discussing a statistical study of electromagnetic ion cyclotron (EMIC) waves detected during 10 years (2001–2010) of Cluster observations. In the first paper, an analysis of EMIC wave properties (i.e., wave power, polarization, normal angle, and wave propagation angle) is presented in both the magnetic latitude (MLAT)-distance as well as magnetic local time (MLT)-L frames. In addition, this paper focuses on the distribution of EMIC wave-associated plasma conditions as well as two EMIC wave generation proxies (the electron plasma frequency to gyrofrequency ratio proxy and the linear theory proxy) in these same frames. Based on the distributions of hot H + anisotropy, electron and hot H+ density measurements, hot H + parallel plasma beta, and the calculated wave generation proxies, three source regions of EMIC waves appear to exist: (1) the well-known overlap between cold plasmaspheric or plume populations with hot anisotropic ring current populations in the postnoon to dusk MLT region; (2) regions all along the dayside magnetosphere at high L shells related to dayside magnetospheric compression and drift shell splitting; and (3) off-equator regions possibly associated with the Shabansky orbits in the dayside magnetosphere.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Kinetic Monte Carlo simulations of structural evolution during anneal of additively manufactured materials

Our experiments indicated that upon a post-processing anneal, an additively manufactured 316L stainless steel exhibits cubic grains rather than the conventional equiaxed grains. In this work, we have used kinetic Monte Carlo simulations to explore the origin of these cubic grains. First, we implemented a new kinetic Monte Carlo model in parallel code SPPARKS to simulate grain growth and recrystallization under a residual energy distribution. Our model incorporates physical properties and real-time, as opposed to generic properties and relative time. We further validated that our SPPARKS simulations reproduced the expected kinetic behavior of single-grain evolution. We then used the validated approach to simulate the anneal of an additively manufactured material under the same conditions used in our experiments. We found that the cubic grains can origin from a periodically varying residual energy that may be present in additively manufactured materials.

36 MATERIALS SCIENCE↗

Computing material volume fractions on a superimposed mesh as applied to Monte Carlo particle transport simulations

Here, we present a newly implemented ray tracing algorithm in OpenMC for efficiently computing material volume fractions on superimposed meshes in complex geometries. By firing rays along each coordinate direction through the geometry, the approach accumulates track-length data in each mesh element, thereby determining the fractional composition of each material. Scaling studies on three different models—a random tetrahedra configuration, the Frascati Neutron Generator ITER dose rate benchmark, and a stellarator design—show excellent parallel performance, with nearly linear speedup on modern multi-threaded and distributed-memory systems. An analysis of the residual error relative to high-resolution reference solutions demonstrated that under optimal conditions it decreases as 1/R, where R is the number of rays fired, making it straightforward to achieve user-prescribed accuracy. This new functionality enables practical, mesh-based approaches for detailed nuclear analyses in production Monte Carlo workflows without resorting to expensive, fully conformal or unstructured meshing.

Monte Carlo↗

Laser-induced slip casting as an additive manufacturing approach for silicon carbide

Here, this work presents processing silicon carbide (SiC) with the laser-induced slip casting (LIS) additive manufacturing (AM). SiC was stabilized in water with polyethyleneimine (PEI) dispersant, and SiC slurries were made with rheology for LIS printing. High-density ceramic parts were printed, followed by single-step binder burnout and sintering. The printed parts achieved 93–95 % of theoretical density. X-ray computed tomography (XCT) revealed a small distribution of flaws exceeding 100 microns. The mechanical properties were measured in both parallel and perpendicular to the printing layers, and the orientation with layers perpendicular to the bending moment resulted in higher strength compared to the parallel direction. Porosity resulting from processing and large inclusions of boron carbide (B4C) were the root cause of failure in the measured samples. Despite these defects through this effort, this new approach demonstrates promise for green forming of SiC with densities greater than 95 % theoretical and tensile strengths above 250 MPa.

Additive Manufacturing↗

High-power, surface-emitting quantum cascade laser operating in a symmetric grating mode

Grating-coupled surface-emitting (GCSE) lasers generally operate with a double-lobed far-field beam pattern along the cavity-length direction, which is a result of lasing being favored in the antisymmetric grating mode. We experimentally demonstrate a GCSE quantum-cascade laser design allowing high-power, nearly single-lobed surface emission parallel to the longitudinal cavity. A 2nd-order Au-semiconductor distributed-feedback (DFB)/distributed-Bragg-reflector (DBR) grating is used for feedback and out-coupling. The DFB and DBR grating regions are 2.55 mm- and 1.28 mm-long, respectively, for a total grating length of 5.1 mm. The lasers are designed to operate in a symmetric (longitudinal) grating mode by causing resonant coupling of the guided optical mode to the antisymmetric surface-plasmon modes of the 2nd-order metal/semiconductor grating. Then, the antisymmetric modes are strongly absorbed by the metal in the grating, causing the symmetric mode to be favored to lase, which, in turn, produces a single-lobed beam over a range of grating duty-cycle values of 36%–41%. Simulations indicate that the symmetric mode is always favored to lase, independent of the random phase of reflections from the device's cleaved ends. Peak pulsed output powers of ∼0.4 W were measured with nearly single-lobe beam-pattern (in the longitudinal direction), single-spatial-mode operation near 4.75 μm wavelength. Far-field measurements confirm a diffraction-limited beam pattern, in agreement with simulations, for a source-to-detector separation of 2 m.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

SUNWARD PROPAGATING ALFVÉN WAVES IN ASSOCIATION WITH SUNWARD DRIFTING PROTON BEAMS IN THE SOLAR WIND

Using measurements from the WIND spacecraft, here we report the observation of sunward propagating Alfvén waves (AWs) in solar wind that is magnetically disconnected from the Earth's bow shock. In the sunward magnetic field sector, we find a period lasting for more than three days in which there existed (during most time intervals) a negative correlation between the flow velocity and magnetic field fluctuations, thus indicating that the related AWs are mainly propagating sunward. Simultaneous observations of counter-streaming suprathermal electrons suggest that these sunward AWs may not simply be due to the deflection of an open magnetic field line. Moreover, no interplanetary coronal mass ejection appears to be associated with the counter-streaming suprathermal electrons. As the scale goes from the magnetohydrodynamic down to the ion kinetic regime, the wave vector of magnetic fluctuations usually becomes more orthogonal to the mean magnetic field direction, and the fluctuations become increasingly compressible, which are both features consistent with quasi-perpendicular kinetic AWs. However, in the case studied here, we find clear signatures of quasi-parallel sunward propagating ion-cyclotron waves. Concurrently, the solar wind proton velocity distribution reveals a sunward field-aligned beam that drifts at about the local Alfvén speed. This beam is found to run in the opposite direction of the normally observed (anti-sunward) proton beam, and is apparently associated with sunward propagating Alfvén/ion-cyclotron waves. The results and conclusions of this study enrich our knowledge of solar wind turbulence and foster our understanding of proton heating and acceleration within a complex magnetic field geometry.

79 ASTRONOMY AND ASTROPHYSICS↗

Grid-based methods for diatomic quantum scattering problems. II. Time-dependent treatment of single- and two-photon ionization of H{sub 2}{sup +}

The time-dependent Schroedinger equation for H{sub 2}{sup +} in a time-varying electromagnetic field is solved in the fixed-nuclei approximation using a previously developed finite-element discrete-variable representation in prolate spheroidal coordinates. Amplitudes for single- and two-photon ionization are obtained using the method of exterior complex scaling to effectively propagate the field-free solutions from the end of the radiation pulse to infinite times. Cross sections are presented for one- and two-photon ionization for both parallel and perpendicular polarizations of the photon field, as well as photoelectron angular distributions for two-photon ionization.

74 ATOMIC AND MOLECULAR PHYSICS↗