Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “parallel performance”

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 685 records · Page 38
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Portable Parallel Algorithms and Frameworks for Exascale Graph Analytics

Graphs (or networks) are a tool used to model the interactions among various entities. Efficiently processing large graphs has recently attracted significant attention due to the applications of graphs in various domains, such as biology, chemistry, and cyber-security. Analyzing the structure and properties of these graphs is an important component of many scientific computing pipelines. With the explosion in the volume of data, graphs have become very large and can contain hundreds of billions of vertices and trillions of edges. Therefore, it is crucial to develop high-performance methods to enable graph analysis to be done quickly and energy-efficiently. Furthermore, these solutions should be highly parallel in order to take advantage of modern parallel machines. However, designing efficient solutions is not enough. With the wide variety of computing environments available, each with different programmability and performance characteristics, it is necessary to develop solutions that are portable in terms of both performance (i.e., provide theoretical guarantees) and programmability (i.e., provide high level abstractions).

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Population Count on Intel® CPU, GPU, and FPGA

Population count is a primitive used in many applications. Commodity processors have dedicated instructions for achieving high-performance population count. Motivated by the productivity of high-level synthesis and the importance of population count, in this paper we investigated the OpenCL implementations of population count algorithms, and evaluated their performance and resource utilizations on an FPGA. Based on the results, we select the most efficient implementation. Then we derived a reduction pattern from a representative application of population count. We parallelized the reduction with atomic functions, and optimized it with vectorized memory accesses, tree reduction, and compute-unit duplication. We evaluated the performance of the reduction kernel on an Intel (R) Xeon (R) CPU and an Intel (R) IrisT Pro integrated GPU, and an FPGA card that features an Intel (R) Arria (R) 10 FPGA. When DRAM memory bandwidth is comparable on the three computing platforms, the FPGA can achieve the highest kernel performance for large workload. On the other hand, we described performance bottlenecks on the FPGA. To make FPGAs more competitive in raw performance compared to high-performant CPU and GPU platforms, it is important to increase external memory bandwidth, minimize data movement between a host and a device, and reduce OpenCL runtime overhead on an FPGA.

heterogeneous computing↗
DOE OSTI2024

Evaluating Performance Portability with the CMS Heterogeneous Pixel Reconstruction code

In the past years the landscape of tools for expressing parallel algorithms in a portable way across various compute accelerators has continued to evolve significantly. There are many technologies on the market that provide portability between CPU, GPUs from several vendors, and in some cases even FPGAs. These technologies include C++ libraries such as Alpaka and Kokkos, compiler directives such as OpenMP, the SYCL open specification that can be implemented as a library or in a compiler, and standard C++ where the compiler is solely responsible for the offloading. Given this developing landscape, users have to choose the technology that best fits their applications and constraints. For example, in the CMS experiment the experience so far in heterogeneous reconstruction algorithms suggests that the full application contains a large number of relatively short computational kernels and memory transfer operations. In this work we use a stand-alone version of the CMS heterogeneous pixel reconstruction code as a realistic use case of HEP reconstruction software that is capable of leveraging GPUs effectively. We summarize the experience of porting this code base from CUDA to Alpaka, Kokkos, SYCL, std::par, and OpenMP offloading. We compare the event processing throughput achieved by each version on NVIDIA and AMD GPUs as well as on a CPU, and compare those to what a native version of the code achieves on each platform.

Andriotis, Nikolaos↗
DOE OSTI2023

Evaluation of OpenAI Codex for HPC Parallel Programming Models Kernel Generation

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.

Godoy, William↗
DOE OSTI2024

Cholla-MHD: An Exascale-capable Magnetohydrodynamic Extension to the Cholla Astrophysical Simulation Code

Abstract We present an extension of the massively parallel, GPU native, astrophysical hydrodynamics code Cholla to magnetohydrodynamics (MHD). Cholla solves the ideal MHD equations in their Eulerian form on a static Cartesian mesh utilizing the Van Leer + constrained transport integrator, the HLLD Riemann solver, and reconstruction methods at second and third order. Cholla’s MHD module can perform ≈260 million cell updates per GPU-second on an NVIDIA A100 while using the HLLD Riemann solver and second order reconstruction. The inherently parallel nature of GPUs combined with increased memory in new hardware allows Cholla’s MHD module to perform simulations with resolutions ∼500 3 cells on a single high-end GPU (e.g., an NVIDIA A100 with 80 GB of memory). We employ GPU direct Message Passing Interface to attain excellent weak scaling on the exascale supercomputer Frontier, while using 74,088 GPUs and simulating a total grid size of over 7.2 trillion cells. A suite of test problems highlights the accuracy of Cholla’s MHD module and demonstrates that zero magnetic divergence in solutions is maintained to round off error. We also present new testing and CI tools using GoogleTest, GitHub Actions, and Jenkins that have made development more robust and accurate and ensure reliability in the future.

Astronomy & Astrophysics↗
DOE OSTI2024

Toward a high-fidelity tritium transport modeling for retention and permeation experiments

We report Tritium Migration Analysis Program version 8 (TMAP8), the latest version of TMAP, was developed within the framework of the Multiphysics Object-Oriented Simulation Environment (MOOSE). Created at Idaho National Laboratory (INL), MOOSE is an open-source, dimension-agnostic, fully coupled, and fully implicit multiphysics platform featuring massively parallel computation capabilities. Using TMAP8, tritium transport in a divertor monoblock was analyzed to elucidate the effects of pulsed operation (up to fifty 1,600 s plasma discharge and cool-down cycles) on the tritium in-vessel inventory source term and ex-vessel release term (i.e., tritium retention and permeation) for safety analysis. With its built-in Message Passing Interface capability, TMAP8 can, in under 2 h, simulate tritium transport in three different layered materials (i.e., tungsten, copper, and copper-chromium-zirconium alloy) in 2D geometry, using a single device/computer with 10 cores. The MOOSE-based TMAP8 code can leverage other MOOSE tools developed under the Nuclear Energy Advanced Modeling and Simulation program to perform tritium and thermal transport in complex geometries and multiphysics environments. And via its massively parallel computation, MOOSE will enable the fusion pilot plant designers to conduct high-fidelity multiphysics modeling for the design of the divertor and blanket systems as well as for the safety analysis.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2024

Neuromorphic overparameterisation and few-shot learning in multilayer physical neural networks

Abstract Physical neuromorphic computing, exploiting the complex dynamics of physical systems, has seen rapid advancements in sophistication and performance. Physical reservoir computing, a subset of neuromorphic computing, faces limitations due to its reliance on single systems. This constrains output dimensionality and dynamic range, limiting performance to a narrow range of tasks. Here, we engineer a suite of nanomagnetic array physical reservoirs and interconnect them in parallel and series to create a multilayer neural network architecture. The output of one reservoir is recorded, scaled and virtually fed as input to the next reservoir. This networked approach increases output dimensionality, internal dynamics and computational performance. We demonstrate that a physical neuromorphic system can achieve an overparameterised state, facilitating meta-learning on small training sets and yielding strong performance across a wide range of tasks. Our approach’s efficacy is further demonstrated through few-shot learning, where the system rapidly adapts to new tasks.

Science & Technology - Other Topics↗
DOE OSTI2024

Simulation-Based Assessment of Energy Consumption of Alternative Powertrains in Agricultural Tractors

The objectives of this research were to develop simulation models for agricultural tractors with different powertrain technologies and evaluate the energy consumption in typical agricultural operations. Simulation models were developed for conventional, parallel hybrid electric, series hybrid electric, fuel cell hybrid, and battery electric powertrains. Autonomie vehicle simulation software (version 2022) was used for the simulations and the tractor models were simulated in two tilling cycles and in a road transport cycle with a trailer. The alternative powertrains were configured to have at least the same tractive performance as the conventional, diesel engine-powered tractor model. The simulation results showed that the potential of the parallel and series hybrid powertrains to improve energy efficiency depends heavily on the tractor size and the operating cycle conditions. The fuel cell hybrid and battery electric powertrains have a higher potential to reduce energy consumption and emissions but still have inherent technical challenges for practical operation. The battery-powered electric tractor would require improvements in the storage energy density to have a comparable operational performance in comparison to other powertrains. The fuel cell hybrid tractor already provided an adequate operating performance but the availability of hydrogen and refueling infrastructure could be challenging to resolve in the farming context.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2021

Calibr8 v.1.0

Calibr8 provides an application to rapidly prototype and perform material model calibration for complex plasticity models using advanced adjoint or forward sensitivity analyses for execution on massively parallel machines. These techniques can be orders of magnitude faster than traditional finite difference approaches for material model calibration. The underlying technology used in Calibr8 is automatic differentiation, which allows for the rapid implementation and testing of new plasticity models within its framework. Additionally, Calibr8 can perform adjoint-based error estimation to approximate discretization errors for user-implemented plasticity models. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525. SAND2021-10630 O

Granzow, BrianN.↗
DOE OSTI2026

Improvements to In-Canal Assay of High Specific Activity 60Co at the ATR

Improvements to the quality of measurements in the assaying of High Specific Activity (HSA) cobalt 60 produced at the Advanced Test Reactor at the Idaho National Laboratory are presented. Such assays are conducted regularly between cycles at the ATR, and due to the radiological safety concerns, they are performed within the ATR canal. A standard ion chamber detector was placed in a collimated assay tube within the canal. Another tube, parallel to the detector, was used to position various cobalt 60 standards and targets. The collimated port between these tubes allows for the targets to be pulled across via a winch system. Previous measurements were performed at discrete 1.5-inch steps using a hand-cranked winch system. Charge was collected and integrated for 10 seconds at each discrete step during a scan. Two substantial improvements to the assay include an electronically controlled, motor-driven winch system and a software controlled continuous charge collection. Combined, these changes improved the quality of measurements as well as streamline the assay process. Further examination of these charge profiles indicated a reasonable amount of photon scattering, or shine, before a line-of-sight is established. This created inaccuracies in the reported activity of the cobalt targets. A series of MCNP-6.3 calculations were performed in an effort to quantify and correct for this shine effect. The results of the calculations and how they were employed to a create shine correction factor for the in-canal cobalt assay at the ATR are presented. The overall bias showed an overestimation of the HSA cobalt 60 activity by around 7-8%.

07 - ISOTOPES AND RADIATION SOURCES↗
DOE OSTI2023

Influence of electronic transport on electrochemical performance of (Cu,Mn) 3 O 4 solid oxide fuel cell cathodes

Alkaline Earth free spinel oxides provide a potential benefit over Sr-doped perovskite-based materials commonly used as electrodes in high-temperature electrochemical energy conversion devices, e.g., solid oxide fuel cells (SOFCs). Sr-segregation is a known issue leading to performance degradation. In this study, Cu x Mn 3-x O 4 (x = 1, 1.2, and 1.5) porous electrodes were examined as SOFC cathodes using electrochemical impedance spectroscopy to investigate the oxygen reduction reaction (ORR) kinetics in relation to the material's intrinsic conductivity, the extrinsic electrode structure, and the cell test design. Similar to the electronic conducting (La,Sr)MnO 3 SOFC cathodes, the ORR kinetics of Cu x Mn 3-x O 4 spinel electrodes was governed by the oxygen adsorption and diffusion at the particle surface as well as the charge transfer at the triple phase boundaries. The overall electrode polarization resistance was highly dependent on contact density with the metallic current collector, active material particle connectivity, electrode thickness, and the intrinsic electronic materials conductivity. Here, we describe the importance of effective electronic charge transport parallel to the electrode surface in maximizing the electrochemically active electrode volume and enhancing electrode performance. We discuss an approach to optimize cell and electrode design with respect to active materials properties. This aspect is critical to ensure reliable evaluation of new materials, since laboratory-scale button-cells typically exhibit a high degree of electrode microstructure (e.g. porosity, thickness) and electrical contact density variation from sample to sample.

(Cu,Mn)3O4↗
DOE OSTI2024

Synthetic β-sheets mimicking fibrillar and oligomeric structures for evaluation of spectral X-ray scattering technique for biomarker quantification

Archetypical cross-β spines sharpen the boundary between functional and pathological proteins including β-amyloid, tau, α-synuclein and transthyretin are linked to many debilitating human neurodegenerative and non-neurodegenerative amyloidoses. An increased focus on development of pathogenic β-sheet specific fluid and imaging structural biomarkers and conformation-specific monoclonal antibodies in targeted therapies has been recently observed. Identification and quantification of pathogenic oligomers remain challenging for existing neuroimaging modalities. We propose two artificial β-sheets which can mimic the nanoscopic structural characteristics of pathogenic oligomers and fibrils for evaluating the performance of a label free, X-ray based biomarker detection and quantification technique. Highly similar structure with elliptical cross-section and parallel cross-β motif is observed among recombinant α-synuclein fibril, Aβ-42 fibril and artificial β-sheet fibrils. We then use these β-sheet models to assess the performance of spectral small angle X-ray scattering (sSAXS) technique for detecting β-sheet structures. sSAXS showed quantitatively accurate detection of antiparallel, cross-β artificial oligomers from a tissue mimicking environment and significant distinction between different oligomer packing densities such as diffuse and dense packings. The proposed synthetic β-sheet models mimicked the nanoscopic structural characteristics of β-sheets of fibrillar and oligomeric states of Aβ and α-synuclein based on the ATR-FTIR and SAXS data. The tunability of β-sheet proportions and shapes of structural motifs, and the low-cost of these β-sheet models can become useful test materials for evaluating β-sheet or amyloid specific biomarkers in a wide range of neurological diseases. By using the proposed synthetic β-sheet models, our study indicates that the sSAXS has potential to evaluate different stages of β-sheet-enriched structures including oligomers of pathogenic proteins.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

Deployment of BISON models of fuel restructuring at high burnup and related fission gas behavior in UO 2

This milestone report details the advancements made in fiscal year 2024 under the Nuclear Energy Advanced Modeling and Simulation (NEAMS) program to improve the modeling of fission gas behavior in high burnup UO 2 nuclear fuel in the BISON fuel performance code. As nuclear fuel is pushed to higher burnups, significant microstructural changes occur within the fuel, including the formation of a high burnup structure (HBS) on the pellet rim and a dark zone deeper within the pellet. These regions, characterized by subgrain formation and increased pore densities, have critical implications for fission gas behavior and release, which are not well understood. The modeling capabilities in BISON did not adequately predict these phenomena, leading to an underestimation of fuel restructuring and - potentially - of fission gas release. To address these gaps, this milestone focused on three key objectives: (1) reviewing and assessing Sifgrs's capabilities for low burnup fuel, on which high burnup capabilities rely, (2) validating and expanding HBS fission gas modeling capabilities, including investigating mechanisms for fission gas release from HBS, and (3) expanding Sifgrs to enable modeling of dark zone formation and its effects on fission gas behavior. These objectives were achieved and are described herein. The achievements of this NEAMS milestone are significant for the industry's goal of burnup extension. The improved predictive modeling capabilities for both low- and high-burnup conditions enhance our understanding of fuel performance under both normal operations and transient scenarios. Although goals were reached, future work is necessary to validate these models against experimental data and quantify their accuracy in different conditions. In parallel, mechanistic modeling efforts should continue to extend and refine these capabilities to increase accuracy while reducing reliance on empirical models. This will ensure robust performance across a broader range of conditions.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2021

Parallels between enzyme catalysis, electrocatalysis, and photoelectrosynthesis

Catalysts are central to accelerating chemistry in biology and technology. In biochemistry, the relationship between the velocity of an enzymatic reaction and the concentration of chemical substrates is described via the Michaelis-Menten model. Additionally, the modeling and benchmarking of synthetic molecular electrocatalysts are also well developed. However, such efforts have not been as rigorously extended to photoelectrosynthetic reactions, where, in addition to chemical substrates and charge carriers, light is a required reagent. In this perspective, we draw parallels between concepts involving enzyme catalytic efficiency, the benchmarking of molecular electrocatalysts, and the performance of photoelectrosynthetic assemblies, while highlighting key differences, assumptions, and limitations.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2021

Reaction Mechanism Generator v3.0: Advances in Automatic Mechanism Generation

In chemical kinetics research, kinetic models containing hundreds of species and tens of thousands of elementary reactions are commonly used to understand and predict the behavior of reactive chemical systems. Reaction Mechanism Generator (RMG) is a software suite developed to automatically generate such models by incorporating and extrapolating from a database of known thermochemical and kinetic parameters. Here, we present the recent version 3 release of RMG and highlight improvements since the previously published description of RMG v1.0. Most notably, RMG can now generate heterogeneous catalysis models in addition to the previously available gas- and liquid-phase capabilities. For model analysis, new methods for local and global uncertainty analysis have been implemented to supplement first-order sensitivity analysis. The RMG database of thermochemical and kinetic parameters has been significantly expanded to cover more types of chemistry. The present release includes parallelization for faster model generation and a new molecule isomorphism approach to improve computational performance. RMG has also been updated to use Python 3, ensuring compatibility with the latest cheminformatics and machine learning packages. Overall, RMG v3.0 includes many changes which improve the accuracy of the generated chemical mechanisms and allow for exploration of a wider range of chemical systems.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Striped Data Analysis Framework

A columnar data representation is known to be an efficient way for data storage, specifically in cases when the analysis is often done based only on a small fragment of the available data structures. A data representation like Apache Parquet is a step forward from a columnar representation, which splits data horizontally to allow for easy parallelization of data analysis. Based on the general idea of columnar data storage, working on the [LDRD Project], we have developed a striped data representation, which, we believe, is better suited to the needs of High Energy Physics data analysis. A traditional columnar approach allows for efficient data analysis of complex structures. While keeping all the benefits of columnar data representations, the striped mechanism goes further by enabling easy parallelization of computations without requiring special hardware. We will present an implementation and some performance characteristics of such a data representation mechanism using a distributed no-SQL database or a local file system, unified under the same API and data representation model. The representation is efficient and at the same time simple so that it allows for a common data model and APIs for wide range of underlying storage mechanisms such as distributed no-SQL databases and local file systems. Striped storage adopts Numpy arrays as its basic data representation format, which makes it easy and efficient to use in Python applications. The Striped Data Server is a web service, which allows to hide the server implementation details from the end user, easily exposes data to WAN users, and allows to utilize well known and developed data caching solutions to further increase data access efficiency. We are considering the Striped Data Server as the core of an enterprise scale data analysis platform for High Energy Physics and similar areas of data processing. We have been testing this architecture with a 2TB dataset from a CMS dark matter search and plan to expand it to multiple 100 TB or even PB scale. We will present the striped format, Striped Data Server architecture and performance test results.

Gutsche, Oliver↗
DOE OSTI2020

DeepFreeze: Towards Scalable Asynchronous Checkpointing of Deep Learning Models

In the age of big data, deep learning has emerged as a powerful tool to extract insight and exploit its value, both in industry and scientific applications. One common pattern emerging in such applications is frequent checkpointing of the state of the learning model during training, needed in a variety of scenarios: analysis of intermediate states to explain features and correlations with training data, exploration strategies involving alternative models that share a common ancestor, knowledge transfer, resilience, etc. However, with increasing size of the learning models and popularity of distributed data-parallel training approaches, simple checkpointing techniques used so far face several limitations: low serialization performance, blocking I/O, stragglers due to the fact that only a single process is involved in checkpointing. This paper proposes a checkpointing technique specifically designed to address the aforementioned limitations, introducing efficient asynchronous techniques to hide the overhead of serialization and I/O, and distribute the load over all participating processes. Experiments with two deep learning applications (CANDLE and ResNet) on a pre-Exascale HPC platform (Theta) shows significant improvement over state-of-art, both in terms of checkpointing duration and runtime overhead.

checkpointing↗
DOE OSTI2024

DC Microgrid Reliability Enhancement with Adaptive Converter Thermal Management

Due to the different device selections, aging levels, and thermal dissipation performance, some converters may take additional thermal stress on switching devices than others in paralleled converter systems, which will reduce system reliability. To address this problem, this paper proposes a power-sharing strategy with adaptive thermal management. First, the temperature-based power loss model and electrical-thermal model are established. Based on that, a high-accuracy IGBT junction temperature estimate considering the power loss-temperature coupling can be achieved. Further, the thermal-sharing for all the switching devices in paralleled converters can be achieved with the proposed adaptive thermal management strategy. The proposed strategy can change the power-sharing ratio adaptively according to the system operation conditions, which will contribute to the system reliability enhancement. The effectiveness of the proposed strategy is verified through PLECS thermal simulation and joint real-time simulation with Dspace and RT-box.

DC microgrid↗