Search NASA⌕ Search

SEARCH · Search NASA

Results for “stencil computation”

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 145 records · Page 8

Numerical simulation of aerothermal loads in hypersonic engine inlets due to shock impingement

The effect of shock impingement on an axial corner simulating the inlet of a hypersonic vehicle engine is modeled using a finite-difference procedure. A three-dimensional dynamic grid adaptation procedure is utilized to move the grids to regions with strong flow gradients. The adaptation procedure uses a grid relocation stencil that is valid at both the interior and boundary points of the finite-difference grid. A linear combination of spatial derivatives of specific flow variables, calculated with finite-element interpolation functions, are used as adaptation measures. This computational procedure is used to study laminar and turbulent Mach 6 flows in the axial corner. The description of flow physics and qualitative measures of heat transfer distributions on cowl and strut surfaces obtained from the analysis are compared with experimental observations. Conclusions are drawn regarding the capability of the numerical scheme for enhanced modeling of high-speed compressible flows.

Ramakrishnan, R.↗

Acoustic Wave Propagation Modeling by a Two-dimensional Finite-difference Summation-by-parts Algorithm

Acoustic waveform modeling is a computationally intensive task and full three-dimensional simulations are often impractical for some geophysical applications such as long-range wave propagation and high-frequency sound simulation. In this study, we develop a two-dimensional high-order accurate finite-difference code for acoustic wave modeling. We solve the linearized Euler equations by discretizing them with the sixth order accurate finite difference stencils away from the boundary and the third order summation-by-parts (SBP) closure near the boundary. Non-planar topographic boundary is resolved by formulating the governing equation in curvilinear coordinates following the interface. We verify the implementation of the algorithm by numerical examples and demonstrate the capability of the proposed method for practical acoustic wave propagation problems in the atmosphere.

58 GEOSCIENCES↗

A real-space Green’s function method for thenumerical solution of Maxwell’s equations

A new method for solving the transverse part of the free-space Maxwell equations in three dimensions is presented. By taking the Helmholtz decomposition of the electric field and current sources and considering only the divergence-free parts, we obtain an explicit real-space representation for the transverse propagator that explicitly respects finite speed of propagation. Because the propagator involves convolution against a singular distribution, we regularize via convolution with smoothing kernels (B-splines) prior to sampling based on a method due to Beyer and LeVeque (1992). We show that the ultimate discrete convolutional propagator can be constructed to attain an arbitrarily high order of accuracy by using higher-order regularizing kernels and finite difference stencils and that it satisfies von Neumann’s stability condition. Furthermore, the propagator is compactly supported and can be applied using Hockney’s method (1970) and parallelized using the same observation as made by Vay, Haber, and Godfrey (2013), leading to a method that is computationally efficient.

Lo, Boris↗

Regional surrogates for predictive control of digital twins

Digital twins of complex systems must involve a model that is fast, generalizable, and usable for real-time control. For example, high-fidelity nonlinear multiphysics simulations can capture laser-material interactions, but are too slow for optimization or model predictive control (MPC). Reduced-order models, used to accelerate such computation, frequently fail to generalize to unseen inputs or control states. We show theoretically that this failure is intrinsic, i.e., that a learned model is non-unique outside the sampled subspace when its low-rank structure arises from limited excitation and clustered eigenvalues, rather than from a user-imposed truncation alone. Motivated by this result, we propose a control-ready regional surrogate-construction framework for both autonomous and nonautonomous dynamics; it employs Koopman lifting to represent nonlinearities, while preserving spatial locality. We illustrate our approach by constructing a control-ready surrogate for the digital twin of a thermal component of additive-manufacturing process. Our surrogate, localized in space through a von Neumann stencil, is learned from noisy high-fidelity simulations that emulate thermal-camera images collected during the manufacturing. It is linear in thermo-physically augmented states so that MPC reduces to a convex quadratic program. The surrogate requires no online correction, generalizes to unseen scan paths and power profiles of the laser, and is more than three orders of magnitude faster than a finite-difference solver. Furthermore, when the MPC sequence computed on the digital twin is applied to this solver, closed-loop temperature regulation is recovered, showing that the surrogate preserves control-relevant input-output behavior.

Data-driven model↗

An Integrated Approach to Scaling Task-Based Runtime Systems for Next Generation Engineering problems

The need to scale next-generation industrial engineering problems to the largest computational platforms presents unique challenges. Such problems may have complex coupled multi-physics models, as well as computationally and communication intensive algorithms not often seen in standard academic problems, while also needing high-demand I/O for analysis purposes. In such cases even codes with good existing scaling properties may need significant changes, addressed in a cross-cutting way to solve such problems at extreme scales. These challenges relate to system components that were not previously problematic on less complex problems and/or at smaller computational scales. This paper illustrates these challenges for Uintah, a highly scalable asynchronous many-task runtime system being applied to the modeling of a 1000 MWe ultra-supercritical coal boiler. In order to model this formidable, production engineering problem, not only was an integrated approach needed that built upon existing scalable components in areas such as complex stencil calculations and linear algebra, but the significant challenges of high demand I/O, complex task graphs, and infrastructure inefficiencies also had to be addressed. This integrated approach allowed this problem to run on 256K CPU cores on Mira, with good weak scaling to 512K CPU cores and similar strong scaling, except for radiation. The need for strong scaling of a new ray tracing-based radiation model required substantial Uintah infrastructure improvements before 119K CPU cores and 7.5K GPUs on Titan could be used, with scaling out to 256K CPU cores and 16K GPUs. The resulting code demonstrates not only excellent overall scalability but a significant improvement in overall performance for the full-scale, production boiler problem.

97 MATHEMATICS AND COMPUTING↗

A Framework for Lattice QCD Calculations on GPUs

Efficient calculation of the greatest common divisor (GCD) for big integers each whose number of bits is greater than or equal to 1024 has drawn a considerable amount of attention because it can be used to detect a weakness of the RSA security infrastructure. This paper presents a parallel binary GCD algorithm and its implementation for big integers on the Intel Xeon Phi coprocessor. This algorithm is capable of computing GCDs efficiently on many pairs of big integers in parallel by utilizing all cores on a Xeon Phi coprocessor as well as taking advantage of all vector processing units of the coprocessor to speed up critical integer operations within the algorithm. Using 240 threads on a Xeon Phi coprocessor to carry out GCD calculations for a large amount of 2048-bit integers, the implementation achieves the speedup of 30 times over a sequential binary GCD algorithm implementation on a single CPU core, and it delivers twice amount of performance in comparison to the same sequential binary GCD implementation running on 240 threads of the Xeon Phi.

Winter, F. T.↗

FFTX-IRIS: Towards Performance Portability and Heterogeneity for SPIRAL Generated Code

FFTX-IRIS is a dynamic system to efficiently utilize novel heterogeneous platforms. This system links two next-generation frameworks, FFTX and IRIS, to navigate the complexity of different hardware architectures. FFTX provides a runtime code generation framework for high-performance Fast Fourier Transform kernels. IRIS runtime provides portability and multi-device heterogeneity, allowing computation on any available compute resource. Together, FFTX-IRIS enables code generation, seamless portability, and performance without user involvement. We show the design of the FFTX-IRIS system along with an evaluation of various small FFT benchmarks. We also demonstrate multi-device heterogeneity of FFTX-IRIS with a larger stencil application.

Rao, Sanil↗

Aspects of a high-resolution scheme for the Navier-Stokes equations

In this paper we emphasize the importance of the form of the numerical dissipation model in computing accurate viscous flow solutions. A high-resolution scheme for viscous flows based on three-point central differencing and a matrix dissipation is considered. The various components of this scheme, including 'entropy fix', limiter function, and boundary-point dissipation are discussed. By analyzing boundary-point dissipation stencils, we confirm that with the matrix dissipation model the normal numerical dissipation terms in the streamwise momentum equation are independent of the Reynolds number. Such independence is not achieved with a scalar dissipation form. The accuracy of the central-difference scheme, with and without matrix dissipation, and the flux-difference split scheme of Roe, which is classified as a high-resolution scheme, is compared. For this comparison, three high Reynolds number laminar flows are considered. Solutions of the Navier-Stokes equations are obtained for low-speed flow over a flat plate, transonic flow over an airfoil with transition near the leading edge, and hypersonic flow over a compression ramp. The emphasis of the comparison is primarily on the details of the viscous flows. The necessity of the high-resolution property is revealed.

Swanson, R. C.↗

A second-order distributed memory parallel fast sweeping method for the Eikonal equation

The Eikonal equation is used to calculate wave propagation and distance fields, and due to its complexity requires numerical treatment for its solution. In this work, we present a second-order distributed memory parallel fast sweeping method. The second-order solution switches on a two-point stencil when two upwind points are available, and reverts to first-order otherwise. In all examples, the second-order method improves the solution over the first-order, allowing for significant savings in memory while achieving the same accuracy. Parallelization over distributed memory saw good weak scaling with optimal convergence. The computational time for second-order was approximately 2.5 times slower than first-order, where the largest amount of mesh points ran on 144 cores (512 GB) was ≈20 billion. The savings in memory from the second-order method combined with the distributed memory algorithm result in the ability to solve problems much larger than are possible with the serial first-order method.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Multi-Dimensional ENO Schemes for General Geometries

A class of ENO schemes is presented for the numerical solution of multidimensional hyperbolic systems of conservation laws in structured and unstructured grids. This is a class of shock-capturing schemes which are designed to compute cell-averages to high order accuracy. The ENO scheme is composed of a piecewise-polynomial reconstruction of the solution form its given cell-averages, approximate evolution of the resulting initial value problem, and averaging of this approximate solution over each cell. The reconstruction algorithm is based on an adaptive selection of stencil for each cell so as to avoid spurious oscillations near discontinuities while achieving high order of accuracy away from them.

Harten, Ami↗

Multi-dimensional Upwind Fluctuation Splitting Scheme with Mesh Adaption for Hypersonic Viscous Flow

A multi-dimensional upwind fluctuation splitting scheme is developed and implemented for two-dimensional and axisymmetric formulations of the Navier-Stokes equations on unstructured meshes. Key features of the scheme are the compact stencil, full upwinding, and non-linear discretization which allow for second-order accuracy with enforced positivity. Throughout, the fluctuation splitting scheme is compared to a current state-of-the-art finite volume approach, a second-order, dual mesh upwind flux difference splitting scheme (DMFDSFV), and is shown to produce more accurate results using fewer computer resources for a wide range of test cases. A Blasius flat plate viscous validation case reveals a more accurate upsilon-velocity profile for fluctuation splitting, and the reduced artificial dissipation production is shown relative to DMFDSFV. Remarkably, the fluctuation splitting scheme shows grid converged skin friction coefficients with only five points in the boundary layer for this case. The second half of the report develops a local, compact, anisotropic unstructured mesh adaptation scheme in conjunction with the multi-dimensional upwind solver, exhibiting a characteristic alignment behavior for scalar problems. The adaptation strategy is extended to the two-dimensional and axisymmetric Navier-Stokes equations of motion through the concept of fluctuation minimization.

Wood, William A., III↗

Preserving Superconvergence of Spectral Elements for Curved Domains [Slides]

Finite Element Methods (FEM) and Spectral Element Methods (SEM) are crucial for solving partial differential equations (PDEs) on complex geometries. SEM offers superior accuracy due to potential superconvergence for simple domains. Challenges persist for domains with curved boundaries, restricting SEM’s advantages in real-world applications. A proposed solution is the introduction of a novel strategy to enhance accuracy and maintain superconvergence of SEM in curved domains. The strategy includes a mesh-generation procedure with geometrically refined elements near curved boundaries and a post-processing phase using the Adaptive Extended Stencil Finite Element Method (AES-FEM). The method, named AES-FEM post-processed Spectral Element Method (ApSEM), aligns the accuracy of non-tensor-product elements with superconvergent spectral elements.

97 MATHEMATICS AND COMPUTING↗

Sharp front tracking with geometric interface reconstruction

Here, this paper presents a novel sharp front-tracking method designed to address limitations in classical front-tracking approaches, specifically their reliance on smooth interpolation kernels and extended stencils for coupling the front and fluid mesh. In contrast, the proposed method employs exclusively sharp, localized interpolation and spreading kernels, restricting the coupling to the interfacial fluid cells–those containing the interface/front. This localized coupling is achieved by integrating a divergence-preserving velocity interpolation method with a piecewise parabolic interface calculation (PPIC) and a polyhedron intersection algorithm to compute the indicator function and local interface curvature. Surface tension is computed using the Continuum Surface Force (CSF) method, maintaining consistency with the sharp representation. Additionally, we propose an efficient local roughness smoothing implementation to account for surface mesh undulations, which is easily applicable to any triangulated surface mesh. Building on our previous work, the primary innovation of this study lies in the localization of the coupling for both the indicator function and surface tension calculations. By reducing the interface thickness on the fluid mesh to a single cell, as opposed to the 4–5 cell spans typical in classical methods, the proposed sharp front-tracking method achieves a highly localized and accurate representation of the interface. This sharper representation mitigates parasitic currents and improves force balancing, making it particularly suitable for scenarios where the interface plays a critical role, such as microfluidics, fluid-fluid interactions, and fluid-structure interactions. The proposed method is comprehensively validated and tested on canonical interfacial flow problems, including stationary and translating Laplace equilibria, oscillating droplets, and rising bubbles. The presented results demonstrate that the sharp front-tracking method significantly outperforms the classical approach in terms of accuracy, stability, and computational efficiency. Notably, parasitic currents are reduced by approximately two orders of magnitude and stable results are obtained for parameter ranges where classical front tracking fails to converge.

42 ENGINEERING↗

The construction of high-accuracy schemes for acoustic equations

An accuracy analysis of various high order schemes is performed from an interpolation point of view. The analysis indicates that classical high order finite difference schemes, which use polynomial interpolation, hold high accuracy only at nodes and are therefore not suitable for time-dependent problems. Thus, some schemes improve their numerical accuracy within grid cells by the near-minimax approximation method, but their practical significance is degraded by maintaining the same stencil as classical schemes. One-step methods in space discretization, which use piecewise polynomial interpolation and involve data at only two points, can generate a uniform accuracy over the whole grid cell and avoid spurious roots. As a result, they are more accurate and efficient than multistep methods. In particular, the Cubic-Interpolated Psuedoparticle (CIP) scheme is recommended for computational acoustics.

Tang, Lei↗

ChatMPI: LLM-Driven MPI Code Generation for HPC Workloads

The Message Passing Interface (MPI) standard plays a crucial role in enabling scientific applications for parallel computing and is an essential component in high-performance computing (HPC). However, implementing MPI code manually—especially applying a proper domain decomposition and communication pattern—is a challenging and error-prone task. We present ChatMPI, an AI assistant for MPI parallelization of sequential C codes. In our analysis, we focus on testing six essential HPC workloads, which are based on Basic Linear Algebra Subprograms levels 1, 2, and 3 as well as sparse, stencil, and iterative operations. We analyze the process of creating ChatMPI by using the ChatHPC library. This lightweight large language model (LLM)–based infrastructure enables HPC experts to efficiently create and supervise trustworthy AI capabilities for critical HPC software tasks. We study the data required for training (fine-tuning) ChatMPI to generate parallel codes that not only use MPI syntax correctly but also apply HPC techniques to reduce memory communication and maximize performance by using proper work decomposition. With a relatively small training dataset composed of a few dozen prompts and fewer than 15 minutes of fine-tuning on one node equipped with two NVIDIA H100 GPUs, ChatMPI elevates trustworthiness for MPI code generation of current LLMs (e.g., Code Llama, ChatGPT-4o and ChatGPT 5). Additionally, we evaluate the performance of the MPI codes generated by ChatMPI in comparison with the ones generated by ChatGPT-4o and ChatGPT-5. The codes generated by ChatMPI provide up to a 4 × boost in performance by using better problem decomposition, communication patterns, and HPC techniques (e.g., communication avoiding).

Valero Lara, Pedro [ORNL] (ORCID:0000000214794310)↗

Algorithmic Extensions of Low-Dispersion Scheme and Modeling Effects for Acoustic Wave Simulation

Accurate computation of acoustic wave propagation may be more efficiently performed when their dispersion relations are considered. Consequently, computational algorithms which attempt to preserve these relations have been gaining popularity in recent years. In the present paper, the extensions to one such scheme are discussed. By solving the linearized, 2-D Euler and Navier-Stokes equations with such a method for the acoustic wave propagation, several issues were investigated. Among them were higher-order accuracy, choice of boundary conditions and differencing stencils, effects of viscosity, low-storage time integration, generalized curvilinear coordinates, periodic series, their reflections and interference patterns from a flat wall and scattering from a circular cylinder. The results were found to be promising en route to the aeroacoustic simulations of realistic engineering problems.

Kaushik, Dinesh K.↗

GPU-accelerated denoising of 3D magnetic resonance images

The raw computational power of GPU accelerators enables fast denoising of 3D MR images using bilateral filtering, anisotropic diffusion, and non-local means. In practice, applying these filtering operations requires setting multiple parameters. This study was designed to provide better guidance to practitioners for choosing the most appropriate parameters by answering two questions: what parameters yield the best denoising results in practice? And what tuning is necessary to achieve optimal performance on a modern GPU? To answer the first question, we use two different metrics, mean squared error (MSE) and mean structural similarity (MSSIM), to compare denoising quality against a reference image. Surprisingly, the best improvement in structural similarity with the bilateral filter is achieved with a small stencil size that lies within the range of real-time execution on an NVIDIA Tesla M2050 GPU. Moreover, inappropriate choices for parameters, especially scaling parameters, can yield very poor denoising performance. To answer the second question, we perform an autotuning study to empirically determine optimal memory tiling on the GPU. The variation in these results suggests that such tuning is an essential step in achieving real-time performance. These results have important implications for the real-time application of denoising to MR images in clinical settings that require fast turn-around times.

97 MATHEMATICS AND COMPUTING↗

Two-step perfectly matched layer for arbitrary-order pseudo-spectral analytical time-domain methods

Numerical simulation of an electrodynamic system in empty space requires the implementation of open boundary conditions (BC) to terminate the solution of Maxwell’s equations on the boundaries of the computational domain. The Perfectly Matched Layer (PML) has become the method of choice for open BC with wave equations, as it is straightforward and relatively easy to implement, and offers very efficient and user-adjustable absorption rates. PMLs are most often employed with the Finite-Difference Time-Domain (FDTD) algorithm, which in its most common implementation offers second-order accuracy in space and time on Cartesian grids. Yet, simulations (including some class of electromagnetic Particle-In-Cell simulations) that require higher precision may resort to higher-order Maxwell solvers employing extended finite-difference stencils, or even to pseudo-spectral Maxwell solvers, for which a general, versatile and efficient formulation of the PML has been missing so far. Here we propose a novel “two-step” formulation of the PML that is simple, very versatile and can be used as is with any Maxwell solver. In particular, it is applicable to a large class of Maxwell solvers including the arbitrary-order Pseudo-Spectral Analytical Time-Domain (PSATD) solver, which offers arbitrarily low numerical dispersion when increasing solver order and becomes dispersion-free at infinite order. Analysis and numerical simulations demonstrate that the new formulation is as efficient as the standard PML formulation, both for the FDTD and the PSATD implementations.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗