SEARCH · Search NASA
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.
Here, stencil codes are compute-intensive algorithms, in which data points arranged in a large grid are being recomputed repeatedly from the values of data points in a predefined neighborhood. This fixed neighborhood pattern is called a stencil. Stencils codes see wide-spread use in computing the discrete solutions of partial differential equations and systems composed of such equations.
This project investigated new techniques for compiling stencil and stencil-like computations. Stencils computations are commonly found in applications such as image processing, physical simulations, image processing, and machine learning. In recent years, many high-performance domain-specific languages (DSLs) have been proposed to optimize stencil computations. To leverage such DSLs, however, existing codes often need to be rewritten. Such rewriting is manual, labor intensive, and error-prone. To alleviate such issues, this project investigated the application of program synthesis and artificial learning techniques to enable stencil computations to automatically leverage new high-performance DSLs. Rather than constructing syntax driven rules, verified lifting uses program synthesis to search for a target code fragment to compile the given input code into. In addition, it also searches for a proof that validates how the found target code fragment preserves the semantics of the original input. Thus, the target code fragment is guaranteed to be semantically equivalent to the input.
In situ visualization and analysis is of increasing importance as the compute and I/O gap further widens with the advance to exascale capable computing. Yet, in situ methods impose resource constraints leading to the difficult task of balancing simulation code performance and the quality of analysis. Applications with tightly-coupled in situ visualization often achieve performance through spatial and temporal downsampling, a tradeoff which risks not capturing transient phenomena at sufficient fidelity. Determining a priori visualization parameters such as sampling rate is difficult without time and resource intensive experimentation. We present a method for reducing resource contention between in situ visualization and stencil codes on heterogeneous systems. This method permits full resolution replay through recording halos and the communication-free reconstruction of interior values uncoupled from the main simulation. We apply this method in the computational fluid dynamics (CFD) code HARVEY [1] on the Summit supercomputer. We demonstrate minimal-overhead, in situ visualization relative to simulation alone, and compare the Halo Replay performance to tightly-coupled in situ approaches.
We evaluate Julia as a single language and ecosystem paradigm powered by LLVM to develop workflow components for high-performance computing. We run a Gray-Scott, 2-variable diffusion-reaction application using a memory-bound, 7-point stencil kernel on Frontier, the US Department of Energy’s first exascale supercomputer. We evaluate the performance, scaling, and trade-offs of (i) the computational kernel on AMD’s MI250x GPUs, (ii) weak scaling up to 4,096 MPI processes/GPUs or 512 nodes, (iii) parallel I/O writes using the ADIOS2 library bindings, and (iv) Jupyter Notebooks for interactive analysis. Results suggest that although Julia generates a reasonable LLVM-IR, a nearly 50% performance difference exists vs. native AMD HIP stencil codes when running on the GPUs. As expected, we observed near-zero overhead when using MPI and parallel I/O bindings for system-wide installed implementations. Consequently, Julia emerges as a compelling high-performance and high-productivity workflow composition language, as measured on the fastest supercomputer in the world.
The high-performance computing (HPC) landscape is quickly changing to systems where most of the performance comes from specialized chips, specifically graphics processing units (GPUs). Such GPU systems are throughput machines, where efficient use of the GPU often requires code refactoring to expose a few orders of magnitude more fine grain parallelism than was previously used on the CPU. Recent modifications to OVERFLOW, an overset, structured grid, computational fluid dynamics flow solver, written in Fortran will be presented. These modifications include both code modernization efforts and algorithmic changes to enable OVERFLOW to efficiently utilize GPUs. Many of these algorithmic changes would likely also be applicable for other structured grid, stencil-based codes wanting to utilize GPUs. The capabilities that have been ported to run on the GPUs are presented, along with the performance gains of the GPU version relative the CPU version of OVERFLOW.
The high-performance computing (HPC) landscape is quickly changing to systems where most of the performance comes from specialized chips, specifically graphics processing units (GPUs). Such GPU systems are throughput machines, where efficient use of the GPU often requires code refactoring to expose a few orders of magnitude more fine grain parallelism than was previously used on the CPU. Recent modifications to OVERFLOW, an overset, structured grid, computational fluid dynamics flow solver, written in Fortran will be presented. These modifications include both code modernization efforts and algorithmic changes to enable OVERFLOW to efficiently utilize GPUs. Many of these algorithmic changes would likely also be applicable for other structured grid, stencil-based codes wanting to utilize GPUs. The capabilities that have been ported to run on the GPUs are presented, along with the performance gains of the GPU version relative the CPU version of OVERFLOW.
A large-eddy simulation (LES) code that utilizes high-resolution numerical schemes is described and applied to a compressible jet flow. The code is written in a general manner such that the accuracy/resolution of the simulation can be selected by the user. Time discretization is performed using a family of low-dispersion Runge-Kutta schemes, selectable from first- to fourth-order. Spatial discretization is performed using central differencing schemes. Both standard schemes, second- to twelfth-order (3 to 13 point stencils) and Dispersion Relation Preserving schemes from 7 to 13 point stencils are available. The code is written in Fortran 90 and uses hybrid MPI/OpenMP parallelization. The code is applied to the simulation of a Mach 0.9 jet flow. Four-stage third-order Runge-Kutta time stepping and the 13 point DRP spatial discretization scheme of Bogey and Bailly are used. The high resolution numerics used allows for the use of relatively sparse grids. Three levels of grid resolution are examined, 3.5, 6.5, and 9.2 million points. Mean flow, first-order turbulent statistics and turbulent spectra are reported. Good agreement with experimental data for mean flow and first-order turbulent statistics is shown.
We derive tight bounds on the cache misses for evaluation of explicit stencil operators on structured grids. Our lower bound is based on the isoperimetrical property of the discrete octahedron. Our upper bound is based on a good surface to volume ratio of a parallelepiped spanned by a reduced basis of the interference lattice of a grid. Measurements show that our algorithm typically reduces the number of cache misses by a factor of three, relative to a compiler optimized code. We show that stencil calculations on grids whose interference lattice have a short vector feature abnormally high numbers of cache misses. We call such grids unfavorable and suggest to avoid these in computations by appropriate padding. By direct measurements on a MIPS R10000 processor we show a good correlation between abnormally high numbers of cache misses and unfavorable three-dimensional grids.
Algorithmic differentiation (AD) is a tool for generating discrete adjoint solvers, which efficiently compute gradients of functions with many inputs, for example for use in gradient-based optimization. AD is often applied to large computations such as stencil operators, which are an important part of most structured-mesh PDE solvers. Stencil computations are often parallelized, for example by using OpenMP, and optimized by using techniques such as cache-blocking and tiling to fully utilize multicore CPUs and many-core accelerators and GPUs. Differentiating these codes with conventional reverse-mode AD results in adjoint codes that cannot be expressed as stencil operations and may not be easily parallelizable. They thus leave most of the compute power of modern architectures unused. We present a method that combines forward-mode AD and loop transformation to generate adjoint solvers that use the same memory access pattern as the original computation that they are derived from and can benefit from the same optimization techniques. The effectiveness of this method is demonstrated by generating a scalable adjoint CFD solver for multicore CPUs and Xeon Phi accelerators.
Stencil computations are not well optimized by general-purpose production compilers and the increased use of multicore, manycore, and accelerator-based systems makes the optimization problem even more challenging. In this paper we present Snowflake, a Domain Specific Language (DSL) for stencils that uses a 'micro-compiler' approach, i.e., small, focused, domain-specific code generators. The approach is similar to that used in image processing stencils, but Snowflake handles the much more complex stencils that arise in scientific computing, including complex boundary conditions, higher-order operators (larger stencils), higher dimensions, variable coefficients, non-unit-stride iteration spaces, and multiple input or output meshes. Snowflake is embedded in the Python language, allowing it to interoperate with popular scientific tools like SciPy and iPython; it also takes advantage of built-in Python libraries for powerful dependence analysis as part of a just-in-time compiler. We demonstrate the power of the Snowflake language and the micro-compiler approach with a complex scientific benchmark, HPGMG, that exercises the generality of stencil support in Snowflake. By generating OpenMP comparable to, and OpenCL within a factor of 2x of hand-optimized HPGMG, Snowflake demonstrates that a micro-compiler can support diverse processor architectures and is performance-competitive whilst preserving a high-level Python implementation.
We evaluate AI-assisted generative capabilities on fundamental numerical kernels in high-performance computing (HPC), including AXPY, GEMV, GEMM, SpMV, Jacobi Stencil, and CG. We test the generated kernel codes for a variety of language-supported programming models, including (1) C++ (e.g., OpenMP [including offload], OpenACC, Kokkos, SyCL, CUDA, and HIP), (2) Fortran (e.g., OpenMP [including offload] and OpenACC), (3) Python (e.g., numpy, Numba, cuPy, and pyCUDA), and (4) Julia (e.g., Threads, CUDA.jl, AMDGPU.jl, and KernelAbstractions.jl). We use the GitHub Copilot capabilities powered by the GPT-based OpenAI Codex available in Visual Studio Code as of April 2023 to generate a vast amount of implementations given simple + + prompt variants. To quantify and compare the results, we propose a proficiency metric around the initial 10 suggestions given for each prompt. Results suggest that the OpenAI Codex outputs for C++ correlate with the adoption and maturity of programming models. For example, OpenMP and CUDA score really high, whereas HIP is still lacking. We found that prompts from either a targeted language such as Fortran or the more general purpose Python can benefit from adding code keywords, while Julia prompts perform acceptably well for its mature programming models (e.g., Threads and CUDA.jl). We expect for these benchmarks to provide a point of reference for each programming model's community. Overall, understanding the convergence of large language models, AI, and HPC is crucial due to its rapidly evolving nature and how it is redefining human-computer interactions.
The SOLPS-ITER code suite is used worldwide for plasma edge modeling, the interpretation of experiments, as well as for the design of the ITER divertor. The numerical scheme of the plasma solver of the code, B2.5, is based on the assumption of perfectly field-aligned grids, while in practice grids are often strongly distorted to match divertor target shapes. Neglecting these grid distortion leads to qualitatively and quantitatively incorrect results for fluid neutral simulations, and may affect results in cold (detached) divertors even when using kinetic neutral simulations. In this contribution, we present the first results of a newly implemented 9-point stencil in B2.5 to properly handle misaligned grids. The new scheme is then applied to fluid neutral simulations of a well-diagnosed and previously modeled Alcator C-Mod discharge. Results are compared with the original 5-point scheme neglecting grid distortion effects, as well as with simulations including a full kinetic neutral model. We conclude that the 9-point stencil is essential to correctly model the transport of fluid neutrals on distorted grids, and to capture the effects of divertor closure on the fluid neutral behavior.
The Charon toolkit for piecemeal development of high-efficiency parallel programs for scientific computing is described. The portable toolkit, callable from C and Fortran, provides flexible domain decompositions and high-level distributed constructs for easy translation of serial legacy code or design to distributed environments. Gradual tuning can subsequently be applied to obtain high performance, possibly by using explicit message passing. Charon also features general structured communications that support stencil-based computations with complex recurrences. Through the separation of partitioning and distribution, the toolkit can also be used for blocking of uni-processor code, and for debugging of parallel algorithms on serial machines. An elaborate review of recent parallelization aids is presented to highlight the need for a toolkit like Charon. Some performance results of parallelizing the NAS Parallel Benchmark SP program using Charon are given, showing good scalability.
The Charon toolkit for piecemeal development of high-efficiency parallel programs for scientific computing is described. The portable toolkit, callable from C and Fortran, provides flexible domain decompositions and high-level distributed constructs for easy translation of serial legacy code or design to distributed environments. Gradual tuning can subsequently be applied to obtain high performance, possibly by using explicit message passing. Charon also features general structured communications that support stencil-based computations with complex recurrences. Through the separation of partitioning and distribution, the toolkit can also be used for blocking of uni-processor code, and for debugging of parallel algorithms on serial machines. An elaborate review of recent parallelization aids is presented to highlight the need for a toolkit like Charon. Some performance results of parallelizing the NAS Parallel Benchmark SP program using Charon are given, showing good scalability. Some performance results of parallelizing the NAS Parallel Benchmark SP program using Charon are given, showing good scalability.
This paper describes an implementation of a high order finite difference technique and its application to the category 2 problems of the ICASE/LaRC Workshop on Computational Aeroacoustics (CAA). Essentially, a popular Computational Fluid Dynamics (CFD) approach (central differencing, Runge-Kutta time integration and artificial dissipation) is modified to handle aeroacoustic problems. The changes include increasing the order of the spatial differencing to sixth order and modifying the artificial dissipation so that it does not significantly contaminate the wave solution. All of the results were obtained from the CM5 located at the Numerical Aerodynamic Simulation Laboratory. lt was coded in CMFortran (very similar to HPF), using programming techniques developed for communication intensive large stencils, and ran very efficiently.
We apply AI-assisted large language model (LLM) capabilities of GPT-3 targeting high-performance computing (HPC) kernels for (i) code generation, and (ii) auto-parallelization of serial code in C ++, Fortran, Python and Julia. Our scope includes the following fundamental numerical kernels: AXPY, GEMV, GEMM, SpMV, Jacobi Stencil, and CG, and language/programming models: (1) C++ (e.g., OpenMP [including offload], OpenACC, Kokkos, SyCL, CUDA, and HIP), (2) Fortran (e.g., OpenMP [including offload] and OpenACC), (3) Python (e.g., numpy, Numba, cuPy, and pyCUDA), and (4) Julia (e.g., Threads, CUDA.jl, AMDGPU.jl, and KernelAbstractions.jl). Kernel implementations are generated using GitHub Copilot capabilities powered by the GPT-based OpenAI Codex available in Visual Studio Code given simple + + prompt variants. To quantify and compare the generated results, we propose a proficiency metric around the initial 10 suggestions given for each prompt. For auto-parallelization, we use ChatGPT interactively giving simple prompts as in a dialogue with another human including simple “prompt engineering” follow ups. Results suggest that correct outputs for C++ correlate with the adoption and maturity of programming models. For example, OpenMP and CUDA score really high, whereas HIP is still lacking. We found that prompts from either a targeted language such as Fortran or the more general-purpose Python can benefit from adding language keywords, while Julia prompts perform acceptably well for its Threads and CUDA.jl programming models. Finally, we expect to provide an initial quantifiable point of reference for code generation in each programming model using a state-of-the-art LLM. Overall, understanding the convergence of LLMs, AI, and HPC is crucial due to its rapidly evolving nature and how it is redefining human-computer interactions.
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.
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).