Search NASA⌕ Search

SEARCH · Search NASA

Results for “CUDA”

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 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Enhancing ChatPORT with CUDA-to-SYCL Kernel Translation Capability

Large Language Models (LLMs) have shown strong capabilities in general code translation. However, code translation involving parallel programming models remains largely unexplored. This work enhances the capabilities of code LLMs in CUDA-to-SYCL kernel translation with parameter-efficient fine-tuning. The resultant fine-tuned LLM, called ChatPORT, is an effort to provide high-fidelity translations from one programming model to another. We describe the preparation of datasets from heterogeneous computing benchmarks for model fine-tuning and testing, the parameter-efficient fine-tuning of 19 open-source code models ranging in size from 0.5 to 34 billion parameters and evaluate the correctness rates of the SYCL kernels by the fine-tuned models. The experimental results show that most code models fail to translate CUDA codes to SYCL correctly. However, fine-tuning these models using a small set of CUDA and SYCL kernels can enhance the capabilities of these models in kernel translation. Depending on the sizes of the models, the correctness rate ranges from 19.9% to 81.7% for a test dataset of 62 CUDA kernels.

Jin, Zheming [ORNL] (ORCID:000000027197780X)↗
NASA NTRS2011

CUDA Simulation of Homogeneous, Incompressible Turbulence

We discuss very fast Compute Unified Device Architecture (CUDA) simulations of ideal homogeneous incompressible turbulence based on Fourier models. These models have associated statistical theories that predict that Fourier coefficients of fluid velocity and magnetic fields (if present) are zero-mean random variables. Prior numerical simulations have shown that certain coefficients have a non-zero mean value that can be very large compared to the associated standard deviation. We review the theoretical basis of this "broken ergodicity" as applied to 2-D and 3-D fluid and magnetohydrodynamic simulations of homogeneous turbulence. Our new simulations examine the phenomenon of broken ergodicity through very long time and large grid size runs performed on a state-of-the-art CUDA platform. Results comparing various CUDA hardware configurations and grid sizes are discussed. NS and MHD results are compared.

Morin, Lee↗
DOE OSTI2025

Accelerating Neutrino Event Generation in MARLEY Using CUDA-Based RNG and GPU Parallelization

MARLEY is a simulation tool that helps scientists study how low-energy neutrinos interact with matter. To work properly, MARLEY uses random numbers thousands of times in each simulation. These random numbers are important for modeling things like how neutrinos collide with atoms and what particles they produce. Right now, MARLEY runs on a regular computer processor (CPU) and uses a built-in random number generator called the Mersenne Twister. This setup works, but it can be slow, especially when trying to simulate many events. This research focuses on making MARLEY run faster by moving the random number generation and some of the repetitive calculations from the CPU to a graphics processing unit (GPU), which can handle many tasks at the same time. We use CUDA (a tool for programming NVIDIA GPUs) and cuRAND (a GPU-based random number library) to test faster alternatives to the current random number system. We compare different GPU-based generators, like curand_mtgp32, xorwow, and philox, to see which ones are the quickest and still give reliable results. Early tests show that using the GPU can make MARLEY simulations much faster. This project not only helps improve current simulation performance but also moves closer to a full simulation chain where all stages can run on modern GPU hardware.

Dunkley, Kimieka [Florida A-M]↗
NASA NTRS2024

Runtime Thread-Block Optimization for Custom Multistream CUDA Kernels for the Glenn Research Center Communication Analysis Suite

In preparation of the return of humans to the Moon with the coming Artemis missions, NASA scientists must evaluate proposed landing site locations for terrain and communications viability. The Glenn Research Center Communication Analysis Suite (GCAS) combines sophisticated communication network models with accurate lunar terrain to access sites across the Moon’s south pole. Given the importance of proper site selection to crew safety and mission success, many locations need to be analyzed resulting in a large computational load needing to be performed. To meet the growing project demands, development has begun to improve the runtime efficiency of GCAS with GPU parallelization by way of multi-stream CUDA kernels. One of the most prominent factors in kernel optimization is the proper selection of thread-block dimensions in order to maximize the concurrent operation on the device. Typically, thread-block dimensions are optimized by hand requiring many stages of benchmarking and iteration. Additionally, given the main conditions to optimization are the physical GPU architecture and problem size, these optimal dimensions are non-portable and fragile in their scope. As such, a novel optimization routine was developed to generate the optimal thread-block dimensions during runtime with considerations to hardware specifications and problem size resolving the issues of portability and enabling the function of more dynamic routines.

Aden Bergstresser↗
NASA NTRS2020

NASA GPU Hackathon Yields Significant Code Improvements

The NASA GPU Hackathon 2020 brought together application developers and computer experts to help get important NASA applications running effectively on graphics processing unit (GPU) nodes. Nine teams of application developers participated in this virtual event, a major impetus for teams to modernize codes of interest for NASA missions to CPU nodes containing GPU accelerators, with a focus on hands-on problem solving. The photo in Figure1 shows 30 of the more than50 participants. The HECC project and NVIDIA jointly organized the event, and HECC provided five Pleiades nodes each with 4 V100 GPUs for teams to use. The virtual event, which took place over four days from September 28–October 7, 2020, used Microsoft Teams and Slack as collaboration tools. Each team consisted of three to six members from NASA Centers and supporting organizations. The teams were paired with one to two mentors from industry, government, and academia. The experience levels of the teams ranged from being GPU novices to advanced CUDA programming experts. OpenACC and the emerging Kokkos API were used in addition to CUDA for GPU programming. During the event, which focused on accelerating AeroSciences and CFD applications, most teams achieved considerable performance improvements on both GPUs and CPUs. For example, a team with no GPU experience completed a first port of a time-critical loop to a GPU. Another team of expert CUDA programmers were able to restructure their algorithm, yielding a factor of five speed-up. And another team sped up some of their CUDA kernels by a factor of 20, which directly translated into their production code. This article highlights some of the many successes resulting from the event.

HECC↗
NASA NTRS2018

C++ Resource Intelligent Compilation for GPU Enabled Applications

We are nearing the limits of Moore's Law with current computing technology. As industries push for more performance from smaller systems, alternate methods of computation such as Graphics Processing Units (GPUs) should be considered. Many of these systems utilize the Compute Unified Device Architecture (CUDA) to give programmers access to individual compute elements of the GPU for general purpose computing tasks. Direct access to the GPU's parallel multi-core architecture enables highly efficient computation and can drastically reduce the time required for complex algorithms or data analysis. Of course not all systems have a CUDA-enabled device to leverage, and so applications must consider optional support for users with these devices. Resource Intelligent Compilation (RIC) addresses this situation by enabling GPU-based acceleration of existing applications without affecting users without GPUs. Resource Intelligent Compilation (RIC) creates C/C++ modules that can be compiled to create a standard CPU version or GPU accelerated version of a program, depending on hardware availability. This is accomplished through a toolbox of programming strategies based on features of the CUDA API. Using this toolbox, existing applications can be modified with ease to support GPU acceleration, and new applications can be generated with just a few simple modifications. All of this culminates in an accelerated application for users with the appropriate hardware, with no performance impact to standard systems. This memorandum presents all the important features involved in supporting and implementing RIC and an example of using RIC to accelerate an existing mathematical model, without removing support for standard users. Through this memorandum, NASA engineers can acquire a set of guidelines to follow for RIC-compliant development, seamlessly accelerating C/C++ applications.

GPU↗
DOE OSTI2024

Portable HCAL reconstruction in the CMS detector using the Alpaka library

CMS has deployed a number of different GPU algorithms at the High-Level Trigger (HLT) in Run 3. As the code base for GPU algorithms continues to grow, the burden for developing and maintaining separate implementations for GPU and CPU becomes increasingly challenging. To mitigate this, CMS has adopted the Alpaka (Abstraction Library for Parallel Kernel Acceleration) library as the performance portability solution to provide a single-code base for parallel execution on both GPUs and CPUs in CMS software (CMSSW). A direct CUDA version of HCAL energy reconstruction, called Minimization At Hcal, Iteratively (MAHI), has been deployed at the HLT in the 2022-2023 data taking period. This contribution will describe how the CUDA version is converted into a portable implementation using the Alpaka library. We will discuss the porting experience from CUDA to Alpaka, the validation process and the performance of the Alpaka version in CPU and GPU.

Kwok, Martin↗
NASA NTRSDate not supplied

Strategies for the GPU Implementation of the OVERFLOW CFD Code

Wondering how to port a large, computational fluid dynamics (CFD) solver, written in Fortran, to run effectively on GPUs? Join this talk to learn about the strategies NASA’s OVERFLOW CFD code has used to effectively utilize GPUs to greatly improve the time to solution compared to CPUs. OVERFLOW is an overset, structured grid, computational fluid dynamics (CFD) flow solver developed by NASA and widely used by government, US industry, and academia. It is known for its effective use of CPU hardware, but this talk will discuss recent efforts to modify the code to run efficiently on GPUs. We will describe our use of OpenACC, CUDA Fortran, and CUDA C++, discussing why and how we use each to map our problem onto NVIDIA GPUs. We will also describe structural changes we made in the code to expose enough parallelism to effectively use the GPU hardware. Finally, we will present the performance benefits from running OVERFLOW on GPUs compared to the well optimized CPU version of the code.

GPU Programming↗
DOE OSTI2024

Large language model evaluation for high–performance computing software development

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.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

A multi-backend autotuning study of feature selection on GPUs

Abstract Feature selection is an important step in machine learning that can benefit from GPU acceleration. As the number of GPU vendors increases, it is imperative to adapt algorithms such as the minimum Redundancy Maximum Relevance (mRMR) feature selection method to different backends that support several GPU architectures. This work presents a multi-backend implementation of mRMR across CUDA, HIP, and SYCL, and studies its performance when combined with Bayesian optimization and transfer learning to automatically tune execution parameters for different platforms and datasets. Our experimental results show that when tuned, CUDA and HIP achieve comparable performance on NVIDIA architectures, while SYCL exhibits a moderate performance gap. Overall, this work highlights the impact of backend choice and autotuning on GPU-accelerated feature selection and provides insights into deploying mRMR across heterogeneous environments.

Beceiro, Bieito (ORCID:0000000333014890)↗
DOE OSTI2025

Mojo: MLIR-based Performance-Portable HPC Science Kernels on GPUs for the Python Ecosystem

We explore the performance and portability of the novel Mojo language for scientific computing workloads on GPUs. As the first language based on the LLVM’s Multi-Level Intermediate Representation (MLIR) compiler infrastructure, Mojo aims to close performance and productivity gaps by combining Python’s interoperability and CUDA-like syntax for compile-time portable GPU programming. We target four scientific workloads: a seven-point stencil (memory-bound), BabelStream (memory-bound), miniBUDE (compute-bound), and Hartree–Fock (compute-bound with atomic operations); and compare their performance against vendor baselines on NVIDIA H100 and AMD MI300A GPUs. We show that Mojo’s performance is competitive with CUDA and HIP for memory-bound kernels, whereas gaps exist on AMD GPUs for atomic operations and for fast-math compute-bound kernels on both AMD and NVIDIA GPUs. Although the learning curve and programming requirements are still fairly low-level, Mojo can close significant gaps in the fragmented Python ecosystem in the convergence of scientific computing and AI.

Godoy, William [ORNL] (ORCID:0000000225905178)↗
DOE OSTI2026

matsim-agents v1.0

matsim-agents is a multi-agent AI framework for atomistic materials simulation and discovery. It orchestrates large language models (LLMs), machine-learned interatomic potentials (MLIPs), and DFT codes into a single agentic loop running on laptops and DOE leadership-class supercomputers. MULTI-AGENT ORCHESTRATION A LangGraph state machine with three nodes: a Planner that converts a natural-language research objective into structured tasks; an Executor that dispatches atomistic tools and loops until the queue is empty; and an Analyst that summarizes results into a human-readable report. State is checkpointed after every step and human-in-the-loop gates can be inserted at any edge. HYPOTHESIS-DRIVEN DISCOVERY CHAT An interactive REPL (matsim-agents chat) that couples LLM dialogue with atomistic simulation. Chemical formulas are automatically detected in conversation turns and trigger a full crystal-phase exploration: structure generation → relaxation → stability scoring → result injection back into the conversation, creating a closed hypothesis-refinement loop. CRYSTAL PHASE ENUMERATION Given a composition, the phase explorer enumerates prototypes by stoichiometry: elemental (fcc/bcc/hcp/sc/diamond), binary 1:1 (rocksalt/CsCl/zincblende/ wurtzite/fluorite/rutile), ternary 1:1:3 (cubic perovskite), ternary 1:2:4 (perovskite + spinel), quaternary 1:1:2:6 (Fm-3m double perovskite). 2-D prototypes (graphene, h-BN, MoS2 2H/1T) and multilayer stacking are also supported via --include-2d and --num-layers. SUPERCELL GENERATION AND SITE DECORATION Auto-tiling to a minimum atom count (--min-atoms), explicit NxNxN tiling (--supercell), symmetry-distinct site decorations (--n-orderings), and isotropic lattice-scale sweeps (--lattice-scales) for volume bracketing. MLFF RELAXATION AND STABILITY SCORING HydraGNN (multi-headed GNN) drives structure relaxation via ASE with FIRE, BFGS, or BFGSLineSearch. Stability output: delta-E/atom ranking across phases and a max-residual-force dynamical-stability proxy. Other MLIPs (MACE, NequIP, Orb) can be plugged in through the same interface. DFT BACKENDS Quantum ESPRESSO pw.x and VASP 6.6 are first-class labellers. Both have validated GPU builds and SLURM/PBS launchers for three DOE platforms: Frontier (AMD MI250X, ROCm), Aurora (Intel PVC, oneAPI), Perlmutter (NVIDIA A100, CUDA). QE produces ~100 binaries (pw.x, ph.x, epw.x, ...). VASP supports scf, relax, vc-relax, and vc-relax-shape run types. ACTIVE-LEARNING LOOP matsim-agents al run CONFIG.yaml drives an iterative HydraGNN-DFT loop: MD generates candidates → ensemble/MC-dropout uncertainty selects the most informative → DFT labels them in parallel inside one allocation → dataset grows → HydraGNN retrains → repeat. DFT backend is a single YAML toggle (dft.backend: vasp | qe). LLM-generated seed structures are supported (no curated POSCAR library needed). Config uses ${VAR}, ${VAR:-default}, ${VAR:?msg} shell-style substitution for cross-user/cross-site portability. LLM BACKENDS Ollama (local, default), vLLM (HPC multi-GPU serving), OpenAI, Anthropic, HuggingFace Transformers+Accelerate. Selected at runtime via flag or env var with no code changes. HPC PORTABILITY Same Python entry points run on Frontier (ROCm 7.2), Aurora (oneAPI), and Perlmutter (CUDA 12). DFT and ML stacks are never co-loaded in the same shell; they couple through the scheduler and filesystem. Advanced multi-node launchers (serve, discovery-chat, single-relaxation, active-learning, QE warm-start) are provided for all three platforms. CODABENCH COMPETITION BUNDLE A self-contained benchmark: 159 atomistic test structures across 11 material classes, 5 tasks (formation energy, forces, ML relaxation, AI-DFT relaxation, phase stability ranking), public/private leaderboard split (30/70), and four ready-to-run baselines: MACE-MP-0, HydraGNN, UMA, AllScAIP.

Lupo Pasini, Massimiliano [Oak Ridge National Labo↗
DOE OSTI2025

Experiences with SYCL on AMD GPUs with Kokkos

With the recent diversification of the hardware landscape in the high-performance computing (HPC) community, performance-portability solutions are becoming more and more important. One of the most popular choices is Kokkos, which recently became a Linux Foundation project. Most of its development is supported by the US Department of Energy and the French Alternative Energies and Atomic Energy Commission. Kokkos is implemented as a C++ library with multiple backends to support CPUs as well as various GPU architectures. These backends include OpenMP, CUDA, HIP, and also SCYL. This approach enables users to leverage the preferred vendor toolchain for the respective platform (e.g. CUDA, ROCm, OneAPI). The SYCL backend is used to target Intel GPUs, in particular to support the Aurora exascale supercomputer. However, SYCL itself also offers a large degree of portability, and in fact Kokkos’ CI for SYCL has been running on NVIDIA hardware due to a lack of access to Intel GPUs. In this report, we describe our experience with using Kokkos SYCL backend on AMD GPUs targeting the Frontier supercomputer at Oak Ridge National Laboratory. The two major SYCL implementations are DPC++ and AdaptiveCpp. While the Kokkos SYCL backend has been implemented using the former, the latter was the first implementation to target AMD GPUs. We will discuss the experience with both of these SYCL implementations in terms of functionality and performance. Using Kokkos to evaluate SYCL toolchains has a number of benefits. Kokkos’ use of SYCL is fairly complex, exercising features such as graphs, relocatable device functions, atomics – including for non-arithmetic types, as well as pinned and page migratable memory allocations. Kokkos also needs to implement capabilities such as Kokkos’ hierarchical parallelism that are not a straight-forward mapping to SYCL capabilities. Furthermore, a large number of libraries and applications that represent diverse use cases are implemented in Kokkos, providing readily available test cases for a toolchain evaluation. Preliminary results show that support for AMD GPUs in DPC++ is much less mature than for NVIDIA GPUs or Intel GPUs. While the situation has improved significantly over the last year, we still encounter many runtime failures, dispatching problems, and code generation issues. With AdaptiveCpp the challenges arise even earlier in the evaluation process. Since Kokkos’ SYCL implementation is largely focused on supporting Intel GPUs, we opted to leverage SYCL extensions which are available in DPC++ but not in AdaptiveCpp. Furthermore, AdaptiveCpp appears to be less conformant with the SYCL2020 standard which Kokkos relies on. In some cases, we are able to work around the lack of feature support, in other cases we have to disable certain Kokkos capabilities to evaluate the toolchain. Our evaluation will leverage Kokkos’ unit tests to establish basic functionality and feature completeness. We then use simple benchmarks for components of a CG implementation as a measure of usability and performance of the SYCL toolchains.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2016

An Optimized Multicolor Point-Implicit Solver for Unstructured Grid Applications on Graphics Processing Units

In the field of computational fluid dynamics, the Navier-Stokes equations are often solved using an unstructuredgrid approach to accommodate geometric complexity. Implicit solution methodologies for such spatial discretizations generally require frequent solution of large tightly-coupled systems of block-sparse linear equations. The multicolor point-implicit solver used in the current work typically requires a significant fraction of the overall application run time. In this work, an efficient implementation of the solver for graphics processing units is proposed. Several factors present unique challenges to achieving an efficient implementation in this environment. These include the variable amount of parallelism available in different kernel calls, indirect memory access patterns, low arithmetic intensity, and the requirement to support variable block sizes. In this work, the solver is reformulated to use standard sparse and dense Basic Linear Algebra Subprograms (BLAS) functions. However, numerical experiments show that the performance of the BLAS functions available in existing CUDA libraries is suboptimal for matrices representative of those encountered in actual simulations. Instead, optimized versions of these functions are developed. Depending on block size, the new implementations show performance gains of up to 7x over the existing CUDA library functions.

Zubair, Mohammad↗
NASA NTRS2016

Fast Simulation of the NICER Instrument

The NICER mission uses a complicated physical system to collect information from objects that are, by x-ray timing science standards, rather faint. To get the most out of the data we will need a rigorous understanding of all instrumental effects. We are in the process of constructing a very fast, high fidelity simulator that will help us to assess instrument performance, support simulation-based data reduction, and improve our estimates of measurement error. We will combine and extend existing optics, detector, and electronics simulations. We will employ the Compute Unied Device Architecture (CUDA2) to parallelize these calculations. The price of suitable CUDA-compatible multi-gigaflop cores is about $0.20/core, so this approach will be very cost-effective.

gEDA↗
NASA NTRSDate not supplied

Performance and Portability of a Linear Solver Across Emerging Architectures

A linear solver algorithm used by a large-scale unstructured-grid computational fluid dynamics application is examined for a broad range of familiar and emerging architectures. Efficient implementation of a linear solver is challenging on recent CPUs offering vector architectures. Vector loads and stores are essential to effectively utilize available memory bandwidth on CPUs, and maintaining performance across different CPUs can be difficult in the face of varying vector lengths offered by each. A similar challenge occurs on GPU architectures, where it is essential to have coalesced memory accesses to utilize memory bandwidth effectively. In this work, we demonstrate that restructuring a computation, and possibly data layout, with regard to architecture is essential to achieve optimal performance by establishing a performance benchmark for each target architecture in a low level language such as vector intrinsics or CUDA. In doing so, we demonstrate how a linear solver kernel can be mapped to Intel® Xeon™ and Xeon Phi™, Marvell® ThunderX2®, NEC® SX-Aurora™ TSUBASA Vector Engine, and NVIDIA® and AMD® GPUs. We further demonstrate that the required code restructuring can be achieved in higher level programming environments such as OpenACC, OCCA, and Intel® OneAPI™/SYCL, and that each generally results in optimal performance on the target architecture. Relative performance metrics for all implementations are shown, and subjective ratings for ease of implementation and optimization are suggested.

Programming models↗