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 91 records · Page 5

Scaling Structured Multigrid to 500K+ Cores Through Coarse-Grid Redistribution

The effcient solution of sparse, linear systems resulting from the discretization of partial differential equations is crucial to the performance of many physics-based simulations. The algorithmic optimality of multilevel approaches for common discretizations makes them a good candidate for an efficient parallel solver. Yet, modern architectures for high-performance computing systems continue to challenge the parallel scalability of multilevel solvers. While algebraic multigrid methods are robust for solving a variety of problems, the increasing importance of data locality and cost of data movement in modern architectures motivates the need to carefully exploit structure in the problem. Robust logically structured variational multigrid methods, such as Black Box Multigrid (BoxMG), maintain structure throughout the multigrid hierarchy. This avoids indirection and increased coarse- grid communication costs typical in parallel algebraic multigrid. Nevertheless, the parallel scalability of structured multigrid is challenged by coarse-grid problems where the overhead in communication dominates computation. In this paper, an algorithm is introduced for redistributing coarse-grid problems through incremental agglomeration. Guided by a predictive performance model, this algorithm provides robust redistribution decisions for structured multilevel solvers. A two-dimensional diffusion problem is used to demonstrate the significant gain in performance of this algorithm over the previous approach that used agglomeration to one processor. In addition, the parallel scalability of this approach

97 MATHEMATICS AND COMPUTING↗

Efficient and Robust Weighted Least-Squares Cell-Average Gradient Construction Methods for the Simulation of Scramjet Flows

The ability to solve the equations governing the hypersonic turbulent flow of a real gas on unstructured grids using a spatially-elliptic, 2nd-order accurate, cell-centered, finite-volume method has been recently implemented in the VULCAN-CFD code. The construction of cell-average gradients using a weighted linear least-squares method and the use of these gradients in the construction of the inviscid fluxes is the focus of this paper. A comparison of least-squares stencil construction methodologies is presented and approaches designed to minimize the number of cells used to augment/stabilize the least-squares stencil while preserving accuracy are explored. Due to our interest in hypersonic flow, a robust multidimensional cell-average gradient limiter procedure that is consistent with the stencil used to construct the cellaverage gradients is described. Canonical problems are computed to illustrate the challenges and investigate the accuracy, robustness and convergence behavior of the cell-average gradient methods on unstructured cell-centered finite-volume grids. Finally, thermally perfect, chemically frozen, Mach 7.8 turbulent flow of air through a scramjet engine flowpath is computed and compared with experimental data to demonstrate the robustness, accuracy and convergence behavior of the preferred gradient method for a realistic 3-D geometry on a non-hex-dominant grid.

White, Jeffery A.↗

A MultiGPU Performance-Portable Solution for Array Programming Based on Kokkos

Today, multiGPU nodes are widely used in high-performance computing and data centers. However, current programming models do not provide simple, transparent, and portable support for automatically targeting multiple GPUs within a node on application areas of array programming. In this paper, we describe a new application programming interface based on the Kokkos programming model to enable array computation on multiple GPUs in a transparent and portable way across both NVIDIA and AMD GPUs. We implement different variations of this technique to accommodate the exchange of stencils (array boundaries) among different GPU memory spaces, and we provide autotuning to select the proper number of GPUs, depending on the computational cost of the operations to be computed on arrays, that is completely transparent to the programmer. We evaluate our multiGPU extension on Summit (#5 TOP500), with six NVIDIA V100 Volta GPUs per node, and Crusher that contains identical hardware/software as Frontier (#1 TOP500), with four AMD MI250X GPUs, each with 2 Graphics Compute Dies (GCDs)for a total of 8 GCDs per node. We also compare the performance of this solution against the use of MPI + Kokkos, which is the cur-rent de facto solution for multiple GPUs in Kokkos. Our evaluation shows that the new Kokkos solution provides good scalability for many GPUs and a faster and simpler solution (from a programming productivity perspective) than MPI + Kokkos.

Valero Lara, Pedro↗

Local search to improve coordinate-based task mapping

We present a local search strategy to improve the coordinate-based mapping of a parallel job’s tasks to the MPI ranks of its parallel allocation in order to reduce network congestion and the job’s communication time. The goal is to reduce the number of network hops between communicating pairs of ranks. Our target is applications with a nearest-neighbor stencil communication pattern running on mesh systems with non-contiguous processor allocation, such as Cray XE and XK Systems. Utilizing the miniGhost mini-app, which models the shock physics application CTH, we demonstrate that our strategy reduces application running time while also reducing the runtime variability. Furthermore, we further show that mapping quality can vary based on the selected allocation algorithm, even between allocation algorithms of similar apparent quality.

97 MATHEMATICS AND COMPUTING↗

A scalable exponential-DG approach for nonlinear conservation laws: With application to Burger and Euler equations

In this work, we propose an Exponential DG framework for partial differential equations. We decompose 7 governing equations into linear and nonlinear parts to which we apply the discontinuous Galerkin 8 (DG) spatial discretization. In particular, we construct the linear part using Jacobian that effectively 9 capture stiff characteristics in the system. The former is integrated analytically, whereas the latter 10 is approximated. This approach i) is stable with a large Courant number (Cr > 1); ii) supports 11 high-order solutions both in time and space; iii) is computationally favorable compared to IMEX 12 DG methods with no preconditioner; iv) becomes comparable to explicit RKDG methods on uniform 13 mesh and beneficial on non-uniform grid for Euler equations; v) is scalable in a modern massively 14 parallel computing architecture due to its explicit nature of exponential time integrators and com15 pact communication stencil of DG method. Numerical results demonstrate the performance of our 16 proposed methods through various examples. We also discuss the stability and convergence analysis 17 for our exponential DG scheme in the context of Burgers equation.

42 ENGINEERING↗

Large-eddy simulations of Richtmyer Meshkov instability in a converging geometry

The Richtmyer-Meshkov instability (RMI) refers to the baroclinic generation of vorticity at a perturbed density interface when impacted by a shock wave. It is often thought of as the impulsive limit of the Rayleigh-Taylor instability. While the RMI has been widely covered in planar geometries, the present simulations investigate the mixing of materials resulting from the interaction of an imploding cylindrical shock wave with a concentric interface, perturbed in both axial and azimuthal directions, which separates outside air from SF{sub 6} (initially 5 times denser) confined in a 90{sup o} wedge. Two incident shocks of Mach numbers M{sub i} = 1.3 and 2.0 at initial impact are tested. These canonical simulations support recent work on understanding the compressible turbulent mixing in converging geometries relevant to both inertial confinement fusion and core-collapse supernova dynamics. Initial irregularities in the density interface form the misalignment between density and pressure gradients required to initiate a first RMI. A second RMI occurs after the initial shock has converged down the wedge, reflected off the axis and reshocks the distorted interface. Reshock interactions of decreasing intensity follow successively. Due to the converging geometry, the accelerated or decelerated motion of the interface also generates Rayleigh-Taylor instabilities. Secondary Kelvin-Helmholtz instabilities develop along the sides of the interpenetrating fingering structures. The energetic reshock produces a large dynamical range of turbulent scales, requiring the utilization of large-eddy simulation (LES). We employed the stretched-vortex subgrid-scale model of turbulent and scalar transport based on an explicit structural modeling of small-scale dynamics. The imploding nature of the flow is particularly suitable for the use of adaptive mesh refinement (AMR) provided by the parallel block-structured AMR framework AMROC. The Favre-filtered Navier-Stokes equations are solved on each Cartesian uniform subgrid of the mesh hierarchy. A weighted, essentially nonoscillatory scheme is used to capture discontinuities but reverts to a low-numerical dissipation, explicit, tuned center-difference stencil in the smooth or turbulent flow regions, optimal for the functioning of our explicit LES. The computational domain is discretized with 95 x 95 x 64 cubic cells on the base grid (64 points in the periodic axial direction), with three additional levels of refinement based on the local density gradient, reducing the computational expenses compared to the equivalent finest unigrid 760 x 760 x 512 problem. The interpenetration of the two fluids after the reshock is clearly discernible in Fig. 1. The mixing is visualized by using isosurfaces of the heavy-fluid mass fraction {psi}. The mixture is colored red as {psi} = 0.90, yellow as {psi} = 0.5, and green as {psi} = 0.1. To illustrate the multiscale nature of the turbulent mixing, we observe that the initial perturbation wavelength is 100 times larger than the finest grid size, which is itself about 100 times larger than the estimated smallest physical scale in the flow - the Kolmogorov scale - where the viscous dissipation of energy occurs. Incident shocks of higher Mach number (e.g., M{sub i} = 2.0) focus the flow further towards the axis, leading to a stronger turbulent mixing.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Towards a Verifiable Domain-Specific Language for Hardware-Accelerated Stencils

Defining a domain-specific language (DSL) that supports vector-calculus abstractions eases the porting of partial differential equation (PDE) solvers to specialized architectures. Sufficiently high-level abstractions empower users to express universal laws with sufficient generality that the laws must always hold true within their domain of validity. A broad class of PDE solvers employs stencil-based algorithms, the target domain of Berkeley Lab's stencil accelerator chip co-design project. First released as open-source in January 2026, the Formal software framework lays a foundation for defining an embedded DSL based on composable operators that implement mimetic numerical methods -- stencil algorithms that guarantee satisfaction of discrete versions of important vector calculus theorems. The Formal DSL will be the frontend to a new class of stencil-PDE accelerators developed jointly by LBNL, UHCL, and UC Berkeley through the DOE Competitive Portfolios for Computer Science Project. This offers the potential of an order of magnitude acceleration for this important category of computational methods to serve the DOE mission. Future work on the Formal DSL will facilitate software verification via type-safe templates that enable problem-specific correctness proofs relying upon generic function theory and carefully crafted unit tests.

Rouson, Damian↗

Direct Replacement of Arbitrary Grid-Overlapping by Non-Structured Grid

A new approach that uses nonstructured mesh to replace the arbitrarily overlapped structured regions of embedded grids is presented. The present methodology uses the Chimera composite overlapping mesh system so that the physical domain of the flowfield is subdivided into regions which can accommodate easily-generated grid for complex configuration. In addition, a Delaunay triangulation technique generates nonstructured triangular mesh which wraps over the interconnecting region of embedded grids. It is designed that the present approach, termed DRAGON grid, has three important advantages: eliminating some difficulties of the Chimera scheme, such as the orphan points and/or bad quality of interpolation stencils; making grid communication in a fully conservative way; and implementation into three dimensions is straightforward. A computer code based on a time accurate, finite volume, high resolution scheme for solving the compressible Navier-Stokes equations has been further developed to include both the Chimera overset grid and the nonstructured mesh schemes. For steady state problems, the local time stepping accelerates convergence based on a Courant - Friedrichs - Leury (CFL) number near the local stability limit. Numerical tests on representative steady and unsteady supersonic inviscid flows with strong shock waves are demonstrated.

Kao, Kai-Hsiung↗

Optimized Combined Compact Differencing Methods for Computational Aeroacoustics

The sixth order Combined Compact Differencing (CCD) scheme of Chu and Fan is optimized, resulting in exceptional performance. Results indicate that the best of these optimized schemes can accurately propagate linear simple harmonic waves with fewer than three points per wavelength, while using a three point differencing stencil. Stable and accurate boundary treatments for these schemes are given, and the performance of these schemes is assessed using a benchmark problem.

Computational Aeroacoustics↗

Implicit fast sweeping method for hyperbolic systems of conservation laws

Implicit time-accurate methods are often used to integrate stiff problems where explicit schemes impose severe time step restrictions. This paper presents an efficient numerical framework based on the Fast Sweeping Method (FSM) for solving linear and nonlinear hyperbolic systems of conservation laws. The solution at each discrete location is computed by sweeping the numerical domain in several predetermined directions that follow the causality of the characteristic families. The use of a fractional step strategy eliminates the need for a solution selection criterion while one-sided stencils limit the number of sweeps to at most 2 d for d space dimensions. This work focuses on the first-order implicit upwind method since it constitutes the building block for high-order conservative schemes. For problems where the degree of stiffness evolves over time, implicit-explicit hybridization can be accomplished with the same algorithm by simply switching the stencil at each time level. As opposed to traditional implicit solvers, the sweeping method does not require a local time linearization of the fluxes thereby preserving the nonlinear stability properties of the original implicit scheme. It also avoids the large computational and memory requirements associated with solving large block-diagonal systems of equations. Here, a series of one- and two-dimensional test cases are presented for the inviscid Burgers' equation and the reactive Euler equations. The results indicate that the implicit FSM can allow a major reduction in the number of time steps even in the presence of discontinuous solution profiles.

74 ATOMIC AND MOLECULAR PHYSICS↗

Compiler-based code generation and autotuning for geometric multigrid on GPU-accelerated supercomputers

GPUs, with their high bandwidths and computational capabilities are an increasingly popular target for scientific computing. Unfortunately, to date, harnessing the power of the GPU has required use of a GPU-specific programming model like CUDA, OpenCL, or OpenACC. Thus, in order to deliver portability across CPU-based and GPU-accelerated supercomputers, programmers are forced to write and maintain two versions of their applications or frameworks. In this paper, we explore the use of a compiler-based autotuning framework based on CUDA-CHiLL to deliver not only portability, but also performance portability across CPU- and GPU-accelerated platforms for the geometric multigrid linear solvers found in many scientific applications. We also show that with autotuning we can attain near Roofline (a performance bound for a computation and target architecture) performance across the key operations in the miniGMG benchmark for both CPU- and GPU-based architectures as well as for a multiple stencil discretizations and smoothers. We show that our technology is readily interoperable with MPI resulting in performance at scale equal to that obtained via hand-optimized MPI+CUDA implementation.

97 MATHEMATICS AND COMPUTING↗

A New Method for the Accurate Calculation of Derivatives at Structured Multiblock Grid Singularities

The new Structured Mapping At Reduced dimension Topology elements (SMART) method for the accurate calculation of derivatives at multiblock structured grid block interfaces is presented. In the SMART method, a ’local’ curvilinear mapping is defined at each interface point, giving a consistent multidimensional definition of the derivatives. The resulting system of equations is overdetermined, and a least squares method is used to determine the value of the derivatives in the local coordinate system. In the special case of a nonsingular interface point, the SMART method automatically recovers the spatial differencing stencils used at the interior points in the neighboring grid blocks.

Computational Aeroacoustics↗

Fresh look at floating shock fitting

A fast implicit upwind procedure for the two-dimensional Euler equations is described that allows accurate computations of shocked flows on nonadapted meshes. Away from shocks, the second-order accurate upwinding is based on the split-coefficient-matrix (SCM) method. In the presence of shocks, the difference stencils are modified using a floating shock fitting technique. Rapid convergence to steady-state solutions is attained with a diagonalized approximate factorization (AF) algorithm. Results are presented for Riemann's problem, for a regular shock reflection at an inviscid wall, for supersonic flow past a cylinder, and for a transonic airfoil. All computed shocks are ideally sharp and in excellent agreement with other numerical results or 'exact' solutions. Most importantly, this has been accomplished on unusually crude meshes without any attempt to align grid lines with shock fronts or to cluster grid lines around shocks.

Hartwich, PETER-M.↗

Patchy nanoparticles by atomic stencilling

Stencilling, in which patterns are created by painting over masks, has ubiquitous applications in art, architecture and manufacturing. Modern, top-down microfabrication methods have succeeded in reducing mask sizes to under 10 nm, enabling ever smaller microdevices as today’s fastest computer chips. Meanwhile, bottom-up masking using chemical bonds or physical interactions has remained largely unexplored, despite its advantages of low cost, solution-processability, scalability and high compatibility with complex, curved and three-dimensional (3D) surfaces. Here we report atomic stencilling to make patchy nanoparticles (NPs), using surface-adsorbed iodide submonolayers to create the mask and ligand-mediated grafted polymers onto unmasked regions as ‘paint’. We use this approach to synthesize more than 20 different types of NP coated with polymer patches in high yield. Polymer scaling theory and molecular dynamics (MD) simulation show that stencilling, along with the interplay of enthalpic and entropic effects of polymers, generates patchy particle morphologies not reported previously. These polymer-patched NPs self-assemble into extended crystals owing to highly uniform patches, including different non-closely packed superlattices. We propose that atomic stencilling opens new avenues in patterning NPs and other substrates at the nanometre length scale, leading to precise control of their chemistry, reactivity and interactions for a wide range of applications, such as targeted delivery, catalysis, microelectronics, integrated metamaterials and tissue engineering.

36 MATERIALS SCIENCE↗

An improved multislope MUSCL scheme for solving shallow water equations on unstructured grids

In this paper we describe an improved vector manipulation multislope monotone upstream-centred scheme for conservation laws (MUSCL) reconstruction for solving the shallow water equations on unstructured grids. This improved MUSCL reconstruction method includes a bigger stencil for the interpolation and saves time for determining the geometric relations compared to the original vector manipulation method, so it is computationally more efficient and straightforward to implement. Four examples involving an analytical solution, laboratory experiments and field-scale measurements are used to test the performance of the proposed scheme. It has been proven that the proposed scheme can provide comparable accuracy and higher efficiency compared to the original vector manipulation method. With the increasing of the number of cells, the advantage of the proposed scheme becomes more apparent.

42 ENGINEERING↗