Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “parallel methods”

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 1,441 records · Page 80
All sourcesNASA NTRSDOE OSTI
DOE OSTI2017

Energy levels, oscillator strengths, line strengths, and transition probabilities in Si-like ions of La XLIII, Er LIV, Tm LV, and Yb LVI

Detailed calculations using the multi-configuration Dirac–Fock (MCDF) method are carried out for the lowest 64 fine-structure levels of the 3s{sup 2}3p{sup 2}, 3s{sup 2}3p3d, 3s3p{sup 3}, 3s3p{sup 2}3d, 3s{sup 2}3d{sup 2}, and 3p{sup 4} configurations in Si-like ions of La XLIII, Er LIV, Tm LV, and Yb LVI. Energies, oscillator strengths, wavelengths, line strengths, and radiative electric dipole transition rates are given for all ions. A parallel calculation using the many-body perturbation theory (MBPT) method is also carried out to assess the present energy levels accuracy. Comparisons are performed between these two sets of energy levels, as well as with other available results, showing that they are in good agreement with each other within 0.5%. These high accuracy results can be used to the modeling and the interpretation of astrophysical objects and fusion plasmas. - Highlights: • Energy levels and E1 transition rates of Si-like ions are presented. • Breit interaction and Quantum Electrodynamics effects are discussed. • Present results should be useful in the astrophysical application and plasma modeling.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗
DOE OSTI2015

SU-C-304-03: Experimental Investigation On the Accuracy of Plastic Scintillation Dosimeters in Small Fields

Purpose: To investigate the accuracy of the Exradin W1 (SI) and of an “in-house” plastic scintillation dosimeter (CHUQ PSD) in small radiation fields. Methods: Output factor (OF) measurements with the W1 and CHUQ PSD were performed for field sizes of 0.5 x 0.5, 1 x 1 and 2 x 2 cm{sup 2}. Both detectors were placed parallel to the central axis (CAX) in water. The spectrum discrimination calibration method was performed in each set-up to account for the Cerenkov (CRV) signal created in the fiber. The OFs were compared to the expected field factors in water derived using i) Monte Carlo (MC) simulations of an accurate accelerator model and ii) microLion (PTW) and D1V diode (SI) OFs. MC-derived correction factors were applied to both the microLion and D1V OFs. For the CHUQ PSD the calibration was repeated in water (// CAX), solid water (perpendicular to CAX) and under a shielded configuration. The signal was collected using a spectrometer (wavelength range = 185–1100 nm). Spectral analysis was performed to evaluate potential changes of the spectral distributions under the various calibration set-up configurations. Results: The W1 OFs presented an over-response for the 0.5 x 0.5 cm{sup 2} in the range of 3 – 4.1% relative to the expected field factor. The CHUQ PSD presented an under-response in the range of 1.5 – 2.7%, without accounting for volume averaging. The CRV spectra under the various calibration procedures appeared similar to each other and only minor changes were observed to the respective OFs. Conclusion: The W1 and CHUQ PSD can be used in small fields down to a 1 x 1 cm{sup 2} field size. Discrepancies were encountered between the two detectors for the smallest field size of 0.5 x 0.5 cm{sup 2} with the CHUQ PSD exhibiting a closer agreement to the expected field factor. Funding sources: 1) Alexander S. Onassis Public Benefit Foundation in Greece and 2) CREATE Medical Physics Research Training Network grant of the Natural Sciences and Engineering Research Council, Grant number: 432290, RGPIN-2014-06475.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
DOE OSTI2018

Computing resonant modes of accelerator cavities by solving nonlinear eigenvalue problems via rational approximation

Here, we present an efficient and reliable algorithm for solving a class of nonlinear eigenvalue problems arising from the modeling of particle accelerator cavities. The eigenvalue nonlinearity in these problems results from the use of waveguides to couple external power sources or to allow certain excited electromagnetic modes to exit the cavity. We use a rational approximation to reduce the nonlinear eigenvalue problem first to a rational eigenvalue problem. We then apply a special linearization procedure to turn the rational eigenvalue problem into a larger linear eigenvalue problem with the same eigenvalues, which can be solved by existing iterative methods. By using a compact scheme to represent both the linearized operator and the eigenvectors to be computed, we obtain a numerical method that only involves solving linear systems of equations of the same dimension as the original nonlinear eigenvalue problem. We refer to this method as a compact rational Krylov (CORK) method. We implemented the CORK method in the Omega3P module of the Advanced Computational Electromagnetic 3D Parallel (ACE3P) simulation suite and validated it by comparing the computed cavity resonant frequencies and damping Q factors of a small model problem to those obtained from a fitting procedure that uses frequency responses computed by another ACE3P module called S3P. We also used the CORK method to compute trapped modes damped in an ideal eight 9-cell SRF cavity cryomodule. This was the first time it was possible to compute these modes directly. The damping Q factors of the computed modes match well with those measured in experiments and the difference in resonant frequencies is within the range introduced by cavity imperfection. Therefore, the CORK method is an extremely valuable tool for computational cavity design.

43 PARTICLE ACCELERATORS↗
NASA NTRS1999

Parallelization of the Physical-Space Statistical Analysis System (PSAS)

Atmospheric data assimilation is a method of combining observations with model forecasts to produce a more accurate description of the atmosphere than the observations or forecast alone can provide. Data assimilation plays an increasingly important role in the study of climate and atmospheric chemistry. The NASA Data Assimilation Office (DAO) has developed the Goddard Earth Observing System Data Assimilation System (GEOS DAS) to create assimilated datasets. The core computational components of the GEOS DAS include the GEOS General Circulation Model (GCM) and the Physical-space Statistical Analysis System (PSAS). The need for timely validation of scientific enhancements to the data assimilation system poses computational demands that are best met by distributed parallel software. PSAS is implemented in Fortran 90 using object-based design principles. The analysis portions of the code solve two equations. The first of these is the "innovation" equation, which is solved on the unstructured observation grid using a preconditioned conjugate gradient (CG) method. The "analysis" equation is a transformation from the observation grid back to a structured grid, and is solved by a direct matrix-vector multiplication. Use of a factored-operator formulation reduces the computational complexity of both the CG solver and the matrix-vector multiplication, rendering the matrix-vector multiplications as a successive product of operators on a vector. Sparsity is introduced to these operators by partitioning the observations using an icosahedral decomposition scheme. PSAS builds a large (approx. 128MB) run-time database of parameters used in the calculation of these operators. Implementing a message passing parallel computing paradigm into an existing yet developing computational system as complex as PSAS is nontrivial. One of the technical challenges is balancing the requirements for computational reproducibility with the need for high performance. The problem of computational reproducibility is well known in the parallel computing community. It is a requirement that the parallel code perform calculations in a fashion that will yield identical results on different configurations of processing elements on the same platform. In some cases this problem can be solved by sacrificing performance. Meeting this requirement and still achieving high performance is very difficult. Topics to be discussed include: current PSAS design and parallelization strategy; reproducibility issues; load balance vs. database memory demands, possible solutions to these problems.

Larson, J. W.↗
DOE OSTI2021

Efficient quadrature rules for finite element discretizations of nonlocal equations

In this paper we design efficient quadrature rules for finite element discretizations of nonlocal diffusion problems with compactly supported kernel functions. Two of the main challenges in nonlocal modeling and simulations are the prohibitive computational cost and the nontrivial implementation of discretization schemes, especially in three-dimensional settings. In this work we circumvent both challenges by introducing a parametrized mollifying function that improves the regularity of the integrand, utilizing an adaptive integration technique, and exploiting parallelization. We first showthat the “mollified” solution converges to the exact one as the mollifying parameter vanishes, then we illustrate the consistency and accuracy of the proposed method on several two- and three-dimensional test cases. Furthermore, we demonstrate the good scaling properties of the parallel implementation of the adaptive algorithm and we compare the proposed method with recently developed techniques for efficient finite element assembly.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS1999

MLP: A Parallel Programming Alternative to MPI for New Shared Memory Parallel Systems

Recent developments at the NASA AMES Research Center's NAS Division have demonstrated that the new generation of NUMA based Symmetric Multi-Processing systems (SMPs), such as the Silicon Graphics Origin 2000, can successfully execute legacy vector oriented CFD production codes at sustained rates far exceeding processing rates possible on dedicated 16 CPU Cray C90 systems. This high level of performance is achieved via shared memory based Multi-Level Parallelism (MLP). This programming approach, developed at NAS and outlined below, is distinct from the message passing paradigm of MPI. It offers parallelism at both the fine and coarse grained level, with communication latencies that are approximately 50-100 times lower than typical MPI implementations on the same platform. Such latency reductions offer the promise of performance scaling to very large CPU counts. The method draws on, but is also distinct from, the newly defined OpenMP specification, which uses compiler directives to support a limited subset of multi-level parallel operations. The NAS MLP method is general, and applicable to a large class of NASA CFD codes.

Taft, James R.↗
NASA NTRS1984

Transition and mixing in impulsively started jets and vortex rings

A new method is used to analyze transition in impulsively started jets and vortex rings. The method uses similarity variables to reduce the equations for unsteady particle paths to a quasi-autonomous system with the Reynolds number as a parameter. The flow pattern is represented in terms of the phase portrait of this system and the structure of the flow is defined by its critical points. Flow transition is examined in terms of bifurcations in the topology of the phase portrait which occur as the Reynolds number is increased. There is no assumption that the flow is parallel or that it is subjected to small disturbances. The method of analysis, through straightforward, is new and lies outside the usual small disturbance theory used to determine the stability of various profile shapes.

Cantwell, B. J.↗
NASA NTRS1992

SAPNEW: Parallel finite element code for thin shell structures on the Alliant FX-80

The finite element method has proven to be an invaluable tool for analysis and design of complex, high performance systems, such as bladed-disk assemblies in aircraft turbofan engines. However, as the problem size increase, the computation time required by conventional computers can be prohibitively high. Parallel processing computers provide the means to overcome these computation time limits. This report summarizes the results of a research activity aimed at providing a finite element capability for analyzing turbomachinery bladed-disk assemblies in a vector/parallel processing environment. A special purpose code, named with the acronym SAPNEW, has been developed to perform static and eigen analysis of multi-degree-of-freedom blade models built-up from flat thin shell elements. SAPNEW provides a stand alone capability for static and eigen analysis on the Alliant FX/80, a parallel processing computer. A preprocessor, named with the acronym NTOS, has been developed to accept NASTRAN input decks and convert them to the SAPNEW format to make SAPNEW more readily used by researchers at NASA Lewis Research Center.

Kamat, Manohar P.↗
NASA NTRS1995

Newton-Krylov-Schwarz: An implicit solver for CFD

Newton-Krylov methods and Krylov-Schwarz (domain decomposition) methods have begun to become established in computational fluid dynamics (CFD) over the past decade. The former employ a Krylov method inside of Newton's method in a Jacobian-free manner, through directional differencing. The latter employ an overlapping Schwarz domain decomposition to derive a preconditioner for the Krylov accelerator that relies primarily on local information, for data-parallel concurrency. They may be composed as Newton-Krylov-Schwarz (NKS) methods, which seem particularly well suited for solving nonlinear elliptic systems in high-latency, distributed-memory environments. We give a brief description of this family of algorithms, with an emphasis on domain decomposition iterative aspects. We then describe numerical simulations with Newton-Krylov-Schwarz methods on aerodynamics applications emphasizing comparisons with a standard defect-correction approach, subdomain preconditioner consistency, subdomain preconditioner quality, and the effect of a coarse grid.

Cai, Xiao-Chuan↗
DOE OSTI2015

Flux tube train model for local turbulence simulation of toroidal plasmas

A new simulation method for local turbulence in toroidal plasmas is developed by extending the conventional idea of the flux tube model. In the new approach, a train of flux tubes is employed, where flux tube simulation boxes are serially connected at each end along a field line so as to preserve a symmetry of the local gyrokinetic equations for image modes in an axisymmetric torus. Validity of the flux tube train model is confirmed against the toroidal ion temperature gradient turbulence for a case with a long parallel correlation of fluctuations, demonstrating numerical advantages over the conventional method in the time step size and the symmetry-preserving property.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2014

Simulation of confined magnetohydrodynamic flows with Dirichlet boundary conditions using a pseudo-spectral method with volume penalization

A volume penalization approach to simulate magnetohydrodynamic (MHD) flows in confined domains is presented. Here the incompressible visco-resistive MHD equations are solved using parallel pseudo-spectral solvers in Cartesian geometries. The volume penalization technique is an immersed boundary method which is characterized by a high flexibility for the geometry of the considered flow. In the present case, it allows to use other than periodic boundary conditions in a Fourier pseudo-spectral approach. The numerical method is validated and its convergence is assessed for two- and three-dimensional hydrodynamic (HD) and MHD flows, by comparing the numerical results with results from literature and analytical solutions. The test cases considered are two-dimensional Taylor–Couette flow, the z-pinch configuration, three dimensional Orszag–Tang flow, Ohmic-decay in a periodic cylinder, three-dimensional Taylor–Couette flow with and without axial magnetic field and three-dimensional Hartmann-instabilities in a cylinder with an imposed helical magnetic field. Finally, we present a magnetohydrodynamic flow simulation in toroidal geometry with non-symmetric cross section and imposing a helical magnetic field to illustrate the potential of the method.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2024

Metallization and stringing for back-contact solar cells

Metallization and stringing methods for back-contact solar cells, and resulting solar cells, are described. In an example, in one embodiment, a method involves aligning conductive wires over the back sides of adjacent solar cells, wherein the wires are aligned substantially parallel to P-type and N-type doped diffusion regions of the solar cells. The method involves bonding the wires to the back side of each of the solar cells over the P-type and N-type doped diffusion regions. The method further includes cutting every other one of the wires between each adjacent pair of the solar cells.

Terao, Akira↗
NASA NTRS1994

Parallel aeroelastic computations for wing and wing-body configurations

The objective of this research is to develop computationally efficient methods for solving fluid-structural interaction problems by directly coupling finite difference Euler/Navier-Stokes equations for fluids and finite element dynamics equations for structures on parallel computers. This capability will significantly impact many aerospace projects of national importance such as Advanced Subsonic Civil Transport (ASCT), where the structural stability margin becomes very critical at the transonic region. This research effort will have direct impact on the High Performance Computing and Communication (HPCC) Program of NASA in the area of parallel computing.

Byun, Chansup↗
NASA NTRS1988

Introduction to a system for implementing neural net connections on SIMD architectures

Neural networks have attracted much interest recently, and using parallel architectures to simulate neural networks is a natural and necessary application. The SIMD model of parallel computation is chosen, because systems of this type can be built with large numbers of processing elements. However, such systems are not naturally suited to generalized communication. A method is proposed that allows an implementation of neural network connections on massively parallel SIMD architectures. The key to this system is an algorithm permitting the formation of arbitrary connections between the neurons. A feature is the ability to add new connections quickly. It also has error recovery ability and is robust over a variety of network topologies. Simulations of the general connection system, and its implementation on the Connection Machine, indicate that the time and space requirements are proportional to the product of the average number of connections per neuron and the diameter of the interconnection network.

Tomboulian, Sherryl↗
NASA NTRS1988

Introduction to a system for implementing neural net connections on SIMD architectures

Neural networks have attracted much interest recently, and using parallel architectures to simulate neural networks is a natural and necessary application. The SIMD model of parallel computation is chosen, because systems of this type can be built with large numbers of processing elements. However, such systems are not naturally suited to generalized elements. A method is proposed that allows an implementation of neural network connections on massively parallel SIMD architectures. The key to this system is an algorithm permitting the formation of arbitrary connections between the neurons. A feature is the ability to add new connections quickly. It also has error recovery ability and is robust over a variety of network topologies. Simulations of the general connection system, and its implementation on the Connection Machine, indicate that the time and space requirements are proportional to the product of the average number of connections per neuron and the diameter of the interconnection network.

Tomboulian, Sherryl↗
DOE OSTI2013

Improving computational efficiency of Monte Carlo simulations with variance reduction

CCFE perform Monte-Carlo transport simulations on large and complex tokamak models such as ITER. Such simulations are challenging since streaming and deep penetration effects are equally important. In order to make such simulations tractable, both variance reduction (VR) techniques and parallel computing are used. It has been found that the application of VR techniques in such models significantly reduces the efficiency of parallel computation due to 'long histories'. VR in MCNP can be accomplished using energy-dependent weight windows. The weight window represents an 'average behaviour' of particles, and large deviations in the arriving weight of a particle give rise to extreme amounts of splitting being performed and a long history. When running on parallel clusters, a long history can have a detrimental effect on the parallel efficiency - if one process is computing the long history, the other CPUs complete their batch of histories and wait idle. Furthermore some long histories have been found to be effectively intractable. To combat this effect, CCFE has developed an adaptation of MCNP which dynamically adjusts the WW where a large weight deviation is encountered. The method effectively 'de-optimises' the WW, reducing the VR performance but this is offset by a significant increase in parallel efficiency. Testing with a simple geometry has shown the method does not bias the result. This 'long history method' has enabled CCFE to significantly improve the performance of MCNP calculations for ITER on parallel clusters, and will be beneficial for any geometry combining streaming and deep penetration effects. (authors)

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2025

Robustness of Deep Learning Classification to Adversarial Input on GPUs: Asynchronous Parallel Accumulation Is a Source of Vulnerability

The ability of machine learning (ML) classification models to resist small, targeted input perturbations—known as adversarial attacks—is a key measure of their safety and reliability. We show that floating-point non associativity (FPNA) coupled with asynchronous parallel programming on GPUs is sufficient to result in misclassification, without any perturbation to the input. Additionally, we show that this misclassification is particularly significant for inputs close to the decision boundary and that standard adversarial robustness results may be overestimated up to 4.6 when not considering machine-level details. We first study a linear classifier, before focusing on standard Graph Neural Network (GNN) architectures and datasets used in robustness assessments. We develop a novel black-box attack using Bayesian optimization to discover external workloads that can change the instruction scheduling which bias the output of reductions on GPUs and reliably lead to misclassification. Motivated by these results, we present a new learnable permutation (LP) gradient-based approach to learning floating-point operation orderings that lead to misclassifications. The LP approach provides a worst-case estimate in a computationally efficient manner, avoiding the need to run identical experiments tens of thousands of times over a potentially large set of possible GPU states or architectures. Finally, using instrumentation-based testing, we investigate parallel reduction ordering across different GPU architectures under external background workloads, when utilizing multi-GPU virtualization, and when applying power capping. Our results demonstrate that parallel reduction ordering varies significantly across architectures under the first two conditions, substantially increasing the search space required to fully test the effects of this parallel scheduler-based vulnerability. These results and the methods developed here can help to include machine-level considerations into adversarial robustness assessments, which can make a difference in safety and mission critical applications.

Shanmugavelu, Sanjif [Maxeler Technologies, a Groq↗
DOE OSTI2018

Analysis of Nonclassical Fracture Problems for Prestressed Bodies with Interacting Cracks

We consider two types of nonclassical fracture mechanisms, namely, the fracture of cracked bodies with initial (residual) stresses acting along the crack planes and the fracture of solids under compression along parallel cracks. To investigate nonaxisymmetric and axisymmetric problems for infinite solids containing two parallel coaxial cracks or a periodic set of coaxial parallel cracks, we use a combined analytic-numerical method within the framework of the three-dimensional linearized mechanics of solids. The analysis involves the representation of stresses and displacements in the linearized theory via harmonic potential functions. With the use of the integral Fourier–Hankel transformations, the problems are reduced to the solution of Fredholm integral equations of the second kind. This approach allows us to investigate problems in a unified general form for compressible and noncompressible homogeneous isotropic or transversely isotropic elastic bodies with an arbitrary structure of the elastic potential, and the material specification of the model is carried out only in the stage of numerical analysis of the resolving equations obtained in the general form. The effects of initial stresses on the stress intensity factors are analyzed for highly elastic materials and layered composites (modeled as transversely isotropic elastic bodies). The “resonance-like” effects are revealed when compressive initial stresses reach the values corresponding to the local loss of stability of the material in the vicinity of cracks, which, according to the indicated combined method, allows one to determine the critical (limiting) load parameters under the conditions of compression of the body along the cracks. The conclusions concerning the dependences of the stress intensity factors and critical (limiting) parameters of compression on the geometric parameters of the problems are formulated as well as concerning the dependences on physical and mechanical characteristics of the materials.

42 ENGINEERING↗