Search NASA⌕ Search

SEARCH · Search NASA

Results for “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 901 records · Page 50
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Efficient Computation of Doppler-Broadened Elastic Scattering Kernel Moments Using Ladder-Operator Formulation

Anefficient routine for computing Legendre moments of the Doppler-broadened elastic scattering kernel, including resonance scattering effects, has been implemented in the ISOXML module of Griffin. Isotropic scattering in the center-of-mass system and the ideal gas model for target motion are assumed. A ladder-operator formulation is introduced to compute all Legendre moments from order 0 to N simultaneously, enabling near-linear scaling of computational cost with respect to the maximum Legendre order. A physics-based strategy for constructing outgoing energy grids has also been developed, in which a tailored base grid is combined with adaptive refinement to maintain accuracy while limiting the number of outgoing energy points. For energies between resonances, a constant cross-section model is employed to further reduce computational cost. In addition, a quantitative criterion is derived to determine isotope-wise cut-off incident energies based on a prescribed up-scattering probability coverage. For 238U, up to incident energies of approximately 75, 230, and 661 eV at 294, 900, and 2500 K (corresponding to a 2% up-scattering probability threshold), computation of P0 kernels requires 1–8 s and computation of P0–P5 kernels requires 0.4–4 min using a single thread, while maintaining 1–3% relative error in up-scattering probability. These results demonstrate that the proposed formulation enables accurate and computationally practical Doppler-broadened kernel generation for online multigroup cross-section production in Griffin.

Doppler-broadening↗
DOE OSTI2018

A sparse matrix–vector multiplication based algorithm for accurate density matrix computations on systems of millions of atoms

Here we present an efficient sparse matrix–vector (SpMV) based method to compute the density matrix P from a given Hamiltonian in electronic structure computations. Our method is a hybrid approach based on Chebyshev–Jackson approximation theory and matrix purification methods like the second order spectral projection purification (SP2). Recent methods to compute the density matrix scale as O(N) in the number of floating point operations but are accompanied by large memory and communication overhead, and they are based on iterative use of the sparse matrix–matrix multiplication kernel (SpGEMM), which is known to be computationally irregular. In addition to irregularity in the sparse Hamiltonian H, the nonzero structure of intermediate estimates of P depends on products of H and evolves over the course of computation. On the other hand, an expansion of the density matrix P in terms of Chebyshev polynomials is straightforward and SpMV based; however, the resulting density matrix may not satisfy the required constraints exactly. In this paper, we analyze the strengths and weaknesses of the Chebyshev–Jackson polynomials and the second order spectral projection purification (SP2) method, and propose to combine them so that the accurate density matrix can be computed using the SpMV computational kernel only, and without having to store the density matrix P. Our method accomplishes these objectives by using the Chebyshev polynomial estimate as the initial guess for SP2, which is followed by using sparse matrix–vector multiplications (SpMVs) to replicate the behavior of the SP2 algorithm for purification. We demonstrate the method on a tight-binding model system of an oxide material containing more than 3 million atoms. In addition, we also present the predicted behavior of our method when applied to near-metallic Hamiltonians with a wide energy spectrum.

42 ENGINEERING↗
DOE OSTI2015

Accelerating population balance-Monte Carlo simulation for coagulation dynamics from the Markov jump model, stochastic algorithm and GPU parallel computing

This paper proposes a comprehensive framework for accelerating population balance-Monte Carlo (PBMC) simulation of particle coagulation dynamics. By combining Markov jump model, weighted majorant kernel and GPU (graphics processing unit) parallel computing, a significant gain in computational efficiency is achieved. The Markov jump model constructs a coagulation-rule matrix of differentially-weighted simulation particles, so as to capture the time evolution of particle size distribution with low statistical noise over the full size range and as far as possible to reduce the number of time loopings. Here three coagulation rules are highlighted and it is found that constructing appropriate coagulation rule provides a route to attain the compromise between accuracy and cost of PBMC methods. Further, in order to avoid double looping over all simulation particles when considering the two-particle events (typically, particle coagulation), the weighted majorant kernel is introduced to estimate the maximum coagulation rates being used for acceptance–rejection processes by single-looping over all particles, and meanwhile the mean time-step of coagulation event is estimated by summing the coagulation kernels of rejected and accepted particle pairs. The computational load of these fast differentially-weighted PBMC simulations (based on the Markov jump model) is reduced greatly to be proportional to the number of simulation particles in a zero-dimensional system (single cell). Finally, for a spatially inhomogeneous multi-dimensional (multi-cell) simulation, the proposed fast PBMC is performed in each cell, and multiple cells are parallel processed by multi-cores on a GPU that can implement the massively threaded data-parallel tasks to obtain remarkable speedup ratio (comparing with CPU computation, the speedup ratio of GPU parallel computing is as high as 200 in a case of 100 cells with 10 000 simulation particles per cell). These accelerating approaches of PBMC are demonstrated in a physically realistic Brownian coagulation case. The computational accuracy is validated with benchmark solution of discrete-sectional method. The simulation results show that the comprehensive approach can attain very favorable improvement in cost without sacrificing computational accuracy.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2017

A boundary integral method for numerical computation of radar cross section of 3D targets using hybrid BEM/FEM with edge elements

This contribution presents the boundary integral formulation for numerical computation of time-harmonic radar cross section for 3D targets. Method relies on hybrid edge element BEM/FEM to compute near field edge element coefficients that are associated with near electric and magnetic fields at the boundary of the computational domain. Special boundary integral formulation is presented that computes radar cross section directly from these edge element coefficients. Consequently, there is no need for near-to-far field transformation (NTFFT) which is common step in RCS computations. By the end of the paper it is demonstrated that the formulation yields accurate results for canonical models such as spheres, cubes, cones and pyramids. Method has demonstrated accuracy even in the case of dielectrically coated PEC sphere at interior resonance frequency which is common problem for computational electromagnetic codes. - Graphical abstract: - Highlights: • Hybrid edge element BEM/FEM is used to find edge element coefficients. • New formulation for RCS computation directly from edge element coefficients is presented. • Formulation is tested on canonical cases and on coated dielectric sphere at interior resonance frequency.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2025

Synthetic-domain computing and neural networks using lithium niobate integrated nonlinear phononics

Analogue computing uses the physical behaviours of devices to provide energy-efficient arithmetic operations. However, scaling up analogue computing platforms by simply increasing the number of devices leads to challenges such as device-to-device variation. Here, in this study, we report scalable analogue computing and neural networks in the synthetic frequency domain using an integrated nonlinear phononic platform on lithium niobate. This synthetic-domain computing is robust to device variations, as vectors and matrices are concurrently encoded at different frequencies within a single device, achieving a high throughput per area. Leveraging inherent nonlinearities, our device-aware neural network can perform a four-class classification task with an accuracy of 98.2%. The nonlinear phononic computing hardware also maintains consistent performance over a wide operational temperature range (characterized up to 192 °C). Our synthetic-domain computing combines single-device parallelism, inherent nonlinearity and environmental stability, and could be of use in edge computing applications in which power efficiency and environmental resilience are crucial.

Ji, Jun [Virginia Polytechnic Inst. and State Univ↗
DOE OSTI2015

Threshold-based queuing system for performance analysis of cloud computing system with dynamic scaling

Cloud computing is promising technology to manage and improve utilization of computing center resources to deliver various computing and IT services. For the purpose of energy saving there is no need to unnecessarily operate many servers under light loads, and they are switched off. On the other hand, some servers should be switched on in heavy load cases to prevent very long delays. Thus, waiting times and system operating cost can be maintained on acceptable level by dynamically adding or removing servers. One more fact that should be taken into account is significant server setup costs and activation times. For better energy efficiency, cloud computing system should not react on instantaneous increase or instantaneous decrease of load. That is the main motivation for using queuing systems with hysteresis for cloud computing system modelling. In the paper, we provide a model of cloud computing system in terms of multiple server threshold-based infinite capacity queuing system with hysteresis and noninstantanuous server activation. For proposed model, we develop a method for computing steady-state probabilities that allow to estimate a number of performance measures.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
DOE OSTI2015

Heterogeneous high throughput scientific computing with APM X-Gene and Intel Xeon Phi

Electrical power requirements will be a constraint on the future growth of Distributed High Throughput Computing (DHTC) as used by High Energy Physics. Performance-per-watt is a critical metric for the evaluation of computer architectures for cost- efficient computing. Additionally, future performance growth will come from heterogeneous, many-core, and high computing density platforms with specialized processors. In this paper, we examine the Intel Xeon Phi Many Integrated Cores (MIC) co-processor and Applied Micro X-Gene ARMv8 64-bit low-power server system-on-a-chip (SoC) solutions for scientific computing applications. As a result, we report our experience on software porting, performance and energy efficiency and evaluate the potential for use of such technologies in the context of distributed computing systems such as the Worldwide LHC Computing Grid (WLCG).

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2021

Sparse Binary Matrix-Vector Multiplication on Neuromorphic Computers

Neuromorphic computers offer the opportunity for low-power, efficient computation. Though they have been primarily applied to neural network tasks, there is also the opportunity to leverage the inherent characteristics of neuromorphic computers (low power, massive parallelism, collocated processing and memory) to perform non-neural network tasks. Here, we demonstrate how an approach for performing sparse binary matrix-vector multiplication on neuromorphic computers. We describe the approach, which relies on the connection between binary matrix-vector multiplication and breadth first search, and we introduce the algorithm for performing this calculation in a neuromorphic way. We validate the approach in simulation. Finally, we provide a discussion of the runtime of this algorithm and discuss where neuromorphic computers in the future may have a computational advantage when performing this computation.

Schuman, Catherine↗
DOE OSTI2024

Evaluation of Autonomous Vehicle Sensing and Compute Load on a Chassis Dynamometer

The sensing and compute load auxiliary energy consumption in autonomous vehicles may be significant due to the large number of sensors and the high compute load from sensor processing and route planning. To understand this issue, this study investigates the top-down energy usage of an electric 2015 Kia Soul fully instrumented with state sensors and a state-specific computer for path planning and sensor processing. A chassis dynamometer was then used to evaluate the cases of (1) no sensors or computation, (2) only sensors operating, and (3) sensors plus compute load. The vehicle was operated autonomously on the dynamometer using a PolySync drive-kit with drive-by-wire longitudinal control. The DynoJet model 224xLC was used to adapt the eddy current dynamometer's road load parameters to comply with an Environmental Protection Agency drive schedule and to evaluate performance against the Argonne National Laboratory Digital Dynamometer Dataset. On the UDDS-HWFET combined driving cycle, the stock battery's range was reduced by 5.6% for sensors alone and 12.2% for sensors and compute load. These results show that the added sensing and compute auxiliary load from automated and autonomous systems is significant and that research efforts need to be spent investigating new energy efficient systems.

Brown, Nicholas E.↗
DOE OSTI2018

Scalable deep text comprehension for Cancer surveillance on high-performance computing

Background: Deep Learning (DL) has advanced the state-of-the-art capabilities in bioinformatics applications which has resulted in trends of increasingly sophisticated and computationally demanding models trained by larger and larger data sets. This vastly increased computational demand challenges the feasibility of conducting cutting-edge research. One solution is to distribute the vast computational workload across multiple computing cluster nodes with data parallelism algorithms. In this study, we used a High-Performance Computing environment and implemented the Downpour Stochastic Gradient Descent algorithm for data parallelism to train a Convolutional Neural Network (CNN) for the natural language processing task of information extraction from a massive dataset of cancer pathology reports. We evaluated the scalability improvements using data parallelism training and the Titan supercomputer at Oak Ridge Leadership Computing Facility. To evaluate scalability, we used different numbers of worker nodes and performed a set of experiments comparing the effects of different training batch sizes and optimizer functions.Results: We found that Adadelta would consistently converge at a lower validation loss, though requiring over twice as many training epochs as the fastest converging optimizer, RMSProp. The Adam optimizer consistently achieved a close 2nd place minimum validation loss significantly faster; using a batch size of 16 and 32 allowed the network to converge in only 4.5 training epochs.Conclusions: We demonstrated that the networked training process is scalable across multiple compute nodes communicating with message passing interface while achieving higher classification accuracy compared to a traditional machine learning algorithm.

Qiu, John X.↗
DOE OSTI2018

A computational fluid dynamics simulation framework for ventricular catheter design optimization

OBJECTIVE Cerebrospinal fluid (CSF) shunts are the primary treatment for patients suffering from hydrocephalus. While proven effective in symptom relief, these shunt systems are plagued by high failure rates and often require repeated revision surgeries to replace malfunctioning components. One of the leading causes of CSF shunt failure is obstruction of the ventricular catheter by aggregations of cells, proteins, blood clots, or fronds of choroid plexus that occlude the catheter’s small inlet holes or even the full internal catheter lumen. Such obstructions can disrupt CSF diversion out of the ventricular system or impede it entirely. Previous studies have suggested that altering the catheter’s fluid dynamics may help to reduce the likelihood of complete ventricular catheter failure caused by obstruction. However, systematic correlation between a ventricular catheter’s design parameters and its performance, specifically its likelihood to become occluded, still remains unknown. Therefore, an automated, open-source computational fluid dynamics (CFD) simulation framework was developed for use in the medical community to determine optimized ventricular catheter designs and to rapidly explore parameter influence for a given flow objective. METHODS The computational framework was developed by coupling a 3D CFD solver and an iterative optimization algorithm and was implemented in a high-performance computing environment. The capabilities of the framework were demonstrated by computing an optimized ventricular catheter design that provides uniform flow rates through the catheter’s inlet holes, a common design objective in the literature. The baseline computational model was validated using 3D nuclear imaging to provide flow velocities at the inlet holes and through the catheter. RESULTS The optimized catheter design achieved through use of the automated simulation framework improved significantly on previous attempts to reach a uniform inlet flow rate distribution using the standard catheter hole configuration as a baseline. While the standard ventricular catheter design featuring uniform inlet hole diameters and hole spacing has a standard deviation of 14.27% for the inlet flow rates, the optimized design has a standard deviation of 0.30%. CONCLUSIONS This customizable framework, paired with high-performance computing, provides a rapid method of design testing to solve complex flow problems. While a relatively simplified ventricular catheter model was used to demonstrate the framework, the computational approach is applicable to any baseline catheter model, and it is easily adapted to optimize catheters for the unique needs of different patients as well as for other fluid-based medical devices.

Neurosciences & Neurology↗
NASA NTRS1974

A Computing Method for Sound Propagation Through a Nonuniform Jet Stream

Understanding the principles of jet noise propagation is an essential ingredient of systematic noise reduction research. High speed computer methods offer a unique potential for dealing with complex real life physical systems whereas analytical solutions are restricted to sophisticated idealized models. The classical formulation of sound propagation through a jet flow was found to be inadequate for computer solutions and a more suitable approach was needed. Previous investigations selected the phase and amplitude of the acoustic pressure as dependent variables requiring the solution of a system of nonlinear algebraic equations. The nonlinearities complicated both the analysis and the computation. A reformulation of the convective wave equation in terms of a new set of dependent variables is developed with a special emphasis on its suitability for numerical solutions on fast computers. The technique is very attractive because the resulting equations are linear in nonwaving variables. The computer solution to such a linear system of algebraic equations may be obtained by well-defined and direct means which are conservative of computer time and storage space. Typical examples are illustrated and computational results are compared with available numerical and experimental data.

Padula, S. L.↗
NASA NTRS1977

Human-computer interaction in multitask situations

Human-computer interaction in multitask decisionmaking situations is considered, and it is proposed that humans and computers have overlapping responsibilities. Queueing theory is employed to model this dynamic approach to the allocation of responsibility between human and computer. Results of simulation experiments are used to illustrate the effects of several system variables including number of tasks, mean time between arrivals of action-evoking events, human-computer speed mismatch, probability of computer error, probability of human error, and the level of feedback between human and computer. Current experimental efforts are discussed and the practical issues involved in designing human-computer systems for multitask situations are considered.

Rouse, W. B.↗
NASA NTRS1978

Computing technology in the 1980's

Advances in computing technology have been led by consistently improving semiconductor technology. The semiconductor industry has turned out ever faster, smaller, and less expensive devices since transistorized computers were first introduced 20 years ago. For the next decade, there appear to be new advances possible, with the rate of introduction of improved devices at least equal to the historic trends. The implication of these projections is that computers will enter new markets and will truly be pervasive in business, home, and factory as their cost diminishes and their computational power expands to new levels. The computer industry as we know it today will be greatly altered in the next decade, primarily because the raw computer system will give way to computer-based turn-key information and control systems.

Stone, H. S.↗
NASA NTRS1978

A single user efficiency measure for evaluation of parallel or pipeline computer architectures

A precise statement of the relationship between sequential computation at one rate, parallel or pipeline computation at a much higher rate, the data movement rate between levels of memory, the fraction of inherently sequential operations or data that must be processed sequentially, the fraction of data to be moved that cannot be overlapped with computation, and the relative computational complexity of the algorithms for the two processes, scalar and vector, was developed. The relationship should be applied to the multirate processes that obtain in the employment of various new or proposed computer architectures for computational aerodynamics. The relationship, an efficiency measure that the single user of the computer system perceives, argues strongly in favor of separating scalar and vector processes, sometimes referred to as loosely coupled processes, to achieve optimum use of hardware.

Jones, W. P.↗
NASA NTRS1978

Comparison of two computer programs by predicting turbulent mixing of helium in a ducted supersonic airstream

Two parabolic flow computer programs, SHIP (a finite-difference program) and COMOC (a finite-element program), are used for predicting three-dimensional turbulent reacting flow fields in supersonic combustors. The theoretical foundation of the two computer programs are described, and then the programs are applied to a three-dimensional turbulent mixing experiment. The cold (nonreacting) flow experiment was performed to study the mixing of helium jets with a supersonic airstream in a rectangular duct. Surveys of the flow field at an upstream were used as the initial data by programs; surveys at a downstream station provided comparison to assess program accuracy. Both computer programs predicted the experimental results and data trends reasonably well. However, the comparison between the computations from the two programs indicated that SHIP was more accurate in computation and more efficient in both computer storage and computing time than COMOC.

Pan, Y. S.↗
NASA NTRS1978

Adaptation of a program for nonlinear finite element analysis to the CDC STAR 100 computer

The conversion of a nonlinear finite element program to the CDC STAR 100 pipeline computer is discussed. The program called DYCAST was developed for the crash simulation of structures. Initial results with the STAR 100 computer indicated that significant gains in computation time are possible for operations on gloval arrays. However, for element level computations that do not lend themselves easily to long vector processing, the STAR 100 was slower than comparable scalar computers. On this basis it is concluded that in order for pipeline computers to impact the economic feasibility of large nonlinear analyses it is absolutely essential that algorithms be devised to improve the efficiency of element level computations.

Pifko, A. B.↗
NASA NTRS1983

A Computer Program for the Calculation of Three-Dimensional Transonic Nacelle/Inlet Flowfields

A highly efficient computer analysis was developed for predicting transonic nacelle/inlet flowfields. This algorithm can compute the three dimensional transonic flowfield about axisymmetric (or asymmetric) nacelle/inlet configurations at zero or nonzero incidence. The flowfield is determined by solving the full-potential equation in conservative form on a body-fitted curvilinear computational mesh. The difference equations are solved using the AF2 approximate factorization scheme. This report presents a discussion of the computational methods used to both generate the body-fitted curvilinear mesh and to obtain the inviscid flow solution. Computed results and correlations with existing methods and experiment are presented. Also presented are discussions on the organization of the grid generation (NGRIDA) computer program and the flow solution (NACELLE) computer program, descriptions of the respective subroutines, definitions of the required input parameters for both algorithms, a brief discussion on interpretation of the output, and sample cases to illustrate application of the analysis.

Vadyak, J.↗