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 649 records · Page 36
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

KinCat v.1.0

SAND2024-02099O The software is designed to allow researchers to perform kinetic Monte Carlo (KMC) simulations of catalytic reactions on a 2D lattice. The code is written efficiently to run on a variety of shared memory computing architectures (e.g. GPU, multi-core) and to natively express the full complexity of lateral interactions on reaction rates. The software allows researchers to perform KMC simulations of catalytic reactions on a 2D lattice. It uses parallel shared-memory computing architectures to reduce run-times and allows for simultaneous simulation of multiple independent runs. 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.

Najm, Habib↗
DOE OSTI2023

Photon Doppler Velocimetry to Spatially Resolve Plasma Density in a Power Flow Gap

The understanding of power flow plasmas is important as we look towards next generation pulsed power (NGPP) as current losses could prohibit the goals of that facility. Therefore, it is important to have accurate diagnostics of the plasma parameters on the current machines, which can be used to help inform and improve simulations. Having these plasma parameters will help validate models and simulations to provide confidence when they are expanded to conditions relevant to NGPP. One important plasma parameter that can be measured is the electron density, which can be measured by photonic Doppler velocimetry (PDV). A PDV system has several key advantages over other interferometers by measuring relatively low densities (> 1 × 10 15 cm -2 ) with both spatial and temporal resolution. Experiments were performed on the Mykonos pulsed power machine, which is a 1 MA sub scale machine in which recent platforms have been developed to explore current densities relevant to the inner magnetically insulated transmission line (MITL) on the Z machine. Experiments were performed on two different platforms, the thin foil platform and the Mykonos parallel plate platform (MP3). In addition, a combination of both single-point and multi-point measurements were used. The single-point measurements proved to be very promising, providing a clear increase in density at about 70 ns into the current rise on thin foil experiments up to about 5 × 10 17 cm -3 before the probe stopped providing signal. While we did also see returns from multi-point measurements on both platforms, the signals were not as easy to interpret due to strong background effects. However, they do show initial promise for this diagnostic to measure density at several points across a 1 mm gap. These measurements provide insights in how to improve the diagnostic so that it can provide useful information on power flow relevant experiments.

47 OTHER INSTRUMENTATION↗
DOE OSTI2024

GASNet-EX Specification Collection (Rev. 2024.5.0)

GASNet-EX is a portable, open-source, high-performance communication library designed to efficiently support the networking requirements of PGAS runtime systems and other alternative models in emerging exascale systems. It provides network-independent, high-performance communication primitives including Remote Memory Access (RMA) and Active Messages (AM). GASNet-EX is an evolution of the popular GASNet communication system, building upon over 20 years of lessons learned, and the primary goals are high performance, interface portability, and expressiveness. The library has been used to implement parallel programming models and libraries such as UPC, UPC++, Fortran coarrays, Legion, Chapel, and many others. This anthology collects together the four separate volumes that currently comprise the GASNet-EX specification, as of the 2024.5.0 release of GASNet-EX.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
DOE OSTI2026

ZEUS: An Efficient GPU Optimization Method Integrating PSO, BFGS, and Automatic Differentiation

We introduce a novel, efficient computational method, ZEUS, for numerical optimization, and provide an open-source implementation. It has four key ingredients: (1) particle swarm optimization (PSO), (2) the use of the Broyden-Fletcher-Goldfarb-Shanno (BFGS) method, (3) automatic differentiation (AD), and (4) GPUs. Our approach addresses the computational challenges inherent in high-dimensional, non-convex optimization problems. In the first phase of the algorithm, we get a potentially good set of starting points using PSO. Thereafter, we run BFGS independently in parallel from these starting points. BFGS is one of the best-performing algorithms for numerical optimization. However, it requires the gradient of the function being optimized. ZEUS integrates automatic differentiation into BFGS thus avoiding the need for the user to calculate derivatives explicitly. The use of GPUs allows ZEUS to speed up the calculations substantially. We carry out systematic studies to explore the trade-offs between the number of PSO iterations taken, starting points, and BFGS iteration depth. We show that a handful of iterations of PSO can improve global convergence when combined with BFGS. We also present performance studies using common test functions. The source code can be found at https://github.com/fnal-numerics/global-optimizer-gpu.

Soos, Dominik [Old Dominion U.]↗
DOE OSTI2022

GPU Acceleration of Large-Scale Full-Frequency GW Calculations

Many-body perturbation theory is a powerful method to simulate electronic excitations in molecules and materials starting from the output of density functional theory calculations. By implementing the theory efficiently so as to run at scale on the latest leadership high-performance computing systems it is possible to extend the scope of GW calculations. Here, we present a GPU acceleration study of the full-frequency GW method as implemented in the WEST code. Excellent performance is achieved through the use of (i) optimized GPU libraries, e.g., cuFFT and cuBLAS, (ii) a hierarchical parallelization strategy that minimizes CPU-CPU, CPU-GPU, and GPU-GPU data transfer operations, (iii) nonblocking MPI communications that overlap with GPU computations, and (iv) mixed precision in selected portions of the code. A series of performance benchmarks has been carried out on leadership high-performance computing systems, showing a substantial speedup of the GPU-accelerated version of WEST with respect to its CPU version. Good strong and weak scaling is demonstrated using up to 25 920 GPUs. Finally, we showcase the capability of the GPU version of WEST for large-scale, full-frequency GW calculations of realistic systems, e.g., a nanostructure, an interface, and a defect, comprising up to 10 368 valence electrons.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

Massively parallel and universal approximation of nonlinear functions using diffractive processors

Nonlinear computation is essential for a wide range of information processing tasks, yet implementing nonlinear functions using optical systems remains a challenge due to the weak and power-intensive nature of optical nonlinearities. Overcoming this limitation without relying on nonlinear optical materials could unlock unprecedented opportunities for ultrafast and parallel optical computing systems. Here, we demonstrate that large-scale nonlinear computation can be performed using linear optics through optimized diffractive processors composed of passive phase-only surfaces. In this framework, the input variables of nonlinear functions are encoded into the phase of an optical wavefront—e.g., via a spatial light modulator (SLM)—and transformed by an optimized diffractive structure with spatially varying point-spread functions to yield output intensities that approximate a large set of unique nonlinear functions–all in parallel. We provide proof establishing that this architecture serves as a universal function approximator for an arbitrary set of bandlimited nonlinear functions, also covering wavelength-multiplexed nonlinear functions as well as multi-variate and complex-valued functions that are all-optically cascadable. Our analysis also indicates the successful approximation of typical nonlinear activation functions commonly used in neural networks, including the sigmoid, tanh, ReLU (rectified linear unit), and softplus. We numerically demonstrate the parallel computation of one million distinct nonlinear functions, accurately executed at wavelength-scale spatial density at the output of a diffractive optical processor. Furthermore, we experimentally validated this framework using in situ optical learning and approximated 35 unique nonlinear functions in a single shot using a compact setup consisting of an SLM and an image sensor. These results establish diffractive optical processors as a scalable platform for massively parallel universal nonlinear function approximation, paving the way for new capabilities in analog optical computing based on linear materials.

Rahman, Md Sadman Sakib [University of California,↗
DOE OSTI2025

MoE-Inference-Bench: Performance Evaluation of Mixture of Expert Large Language and Vision Models

Mixture of Experts (MoE) models have enabled the scaling of Large Language Models (LLMs) and Vision Language Models (VLMs) by achieving massive parameter counts while maintaining computational efficiency. However, MoEs introduce several inference-time challenges, including load imbalance across experts and the additional routing computational overhead. To address these challenges and fully harness the benefits of MoE, a systematic evaluation of hardware acceleration techniques is essential. We present MoE-Inference-Bench, a comprehensive study to evaluate MoE performance across diverse scenarios. We analyze the impact of batch size, sequence length, and critical MoE hyperparameters such as FFN dimensions and number of experts on throughput. We evaluate several optimization techniques on Nvidia H100 GPUs, including pruning, Fused MoE operations, speculative decoding, quantization, and various parallelization strategies. Our evaluation includes MoEs from the Mixtral, DeepSeek, OLMoE and Qwen families. The results reveal performance differences across configurations and provide insights for the efficient deployment of MoEs.

Chitty-Venkata, Krishna Teja↗
DOE OSTI2024

Evolution of the SLATE linear algebra library

SLATE (Software for Linear Algebra Targeting Exascale) is a distributed, dense linear algebra library targeting both CPU-only and GPU-accelerated systems, developed over the course of the Exascale Computing Project (ECP). While it began with several documents setting out its initial design, significant design changes occurred throughout its development. In some cases, these were anticipated: an early version used a simple consistency flag that was later replaced with a full-featured consistency protocol. In other cases, performance limitations and software and hardware changes prompted a redesign. Sequential communication tasks were parallelized; host-to-host MPI calls were replaced with GPU device-to-device MPI calls; more advanced algorithms such as Communication Avoiding LU and the Random Butterfly Transform (RBT) were introduced. Early choices that turned out to be cumbersome, error prone, or inflexible have been replaced with simpler, more intuitive, or more flexible designs. Applications have been a driving force, prompting a lighter weight queue class, nonuniform tile sizes, and more flexible MPI process grids. Of paramount importance has been building a portable library that works across several different GPU architectures – AMD, Intel, and NVIDIA – while keeping a clean and maintainable codebase. Here we explore the evolving design choices and their effects, both in terms of performance and software sustainability.

Gates, Mark↗
DOE OSTI2022

Fast External Quantum Efficiency Measurements of Photovoltaic Modules

The external quantum efficiency (EQE) or associated spectral response is widely used to evaluate the performance of photovoltaics, light emitting diodes, photodetectors, photodiodes, semiconductor lasers, laser-induced refrigeration of solids etc. For photovoltaic (PV) devices, The EQE is used to determine the spectral mismatch correction that is used to convert a performance measurement from a non-ideal simulator spectrum to a standard spectrum in PV device calibrations. Moreover, EQE can provide important insight into the physics of a PV device. For instance, the EQE shape can be used to infer mechanisms limiting the performance of the device. However, EQE measurements of full-size PV modules remain a challenge in the PV community due to rare availability of reliable module QE measurement tools worldwide and the size and complexity of module architecture that comprises several cells connected in strings (series or/and parallel). Here, we use a customized nondestructive tool made of high-power light emitting diodes to perform high-throughput EQE measurements on PV modules. Circuit simulations show that the measurement system can safely be used to measure both crystalline silicon and thin film-based PV modules with series connected cells without the risk of damage through reverse bias. In addition, EQE mapping combined with electroluminescence show a strong correlation and provide a way to determine both spatial non-uniformity of the spectral response and defects/degradation effects across the module.

PV module↗
DOE OSTI2022

VTK-m User's Guide (V.1.8)

Extreme scale systems require a new programming model and a fundamental change in how we design algorithms. To address these issues we created VTK-m: the visualization toolkit for multi-/many-core architectures. VTK-m supports a number of algorithms and the ability to design further algorithms through a top-down design with an emphasis on extreme parallelism. VTK-m also provides support for finding and building links across topologies, making it possible to perform operations that determine manifold surfaces, interpolate generated values, and find adjacencies. Although VTK-m provides a simplified high-level interface for programming, its template-based code removes the overhead of abstraction. VTK-m simplifies the development of parallel scientific visualization algorithms by providing a framework of supporting functionality that allows developers to focus on visualization operations. Additionally, VTK-m provides data abstractions not provided by other libraries that make code written in VTK-m more versatile.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Internal Standard Triggered-Parallel Reaction Monitoring Mass Spectrometry Enables Multiplexed Quantification of Candidate Biomarkers in Plasma

Despite advances in proteomic technologies, clinical translation of plasma biomarkers remains low, partly due to a major bottleneck between the discovery of candidate biomarkers and costly clinical validation studies. Due to a dearth of multiplexable assays, generally only a few candidate biomarkers are tested, and the validation success rate is accordingly low. Previously, mass spectrometry-based approaches have been used to fill this gap but feature poor quantitative performance and were generally limited to hundreds of proteins. Here, we demonstrate the capability of an internal standard triggered-parallel reaction monitoring (IS-PRM) assay to greatly expand the numbers of candidates that can be tested with improved quantitative performance. The assay couples immunodepletion and fractionation with IS-PRM and was developed and implemented in human plasma to quantify 5176 peptides representing 1314 breast cancer biomarker candidates. Characterization of the IS-PRM assay demonstrated the precision (median % CV of 7.7%), linearity (median R 2 > 0.999 over 4 orders of magnitude), and sensitivity (median LLOQ < 1 fmol, approximately) to enable rank-ordering of candidate biomarkers for validation studies. Using three plasma pools from breast cancer patients and three control pools, 893 proteins were quantified, of which 162 candidate biomarkers were verified in at least one of the cancer pools and 22 were verified in all three cancer pools. The assay greatly expands capabilities for quantification of large numbers of proteins and is well suited for prioritization of viable candidate biomarkers.

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

MPACT Verification With Magnox Reactor Neutronics Progression Problems

MPACT is a state-of-the-art core simulator designed to perform high-fidelity analysis using whole-core, three-dimensional, pin-resolved neutron transport calculations on modern parallel computing hardware. MPACT was originally developed to model light water reactors, and its capabilities are being extended to simulate gas-cooled, graphite-moderated cores such as Magnox reactors. To verify MPACT’s performance in this new application, the code is being formally benchmarked using representative problems. Progression problems are a series of example models that increase in complexity designed to test a code’s performance. The progression problems include both beginning-of-cycle and depletion calculations. Reference solutions for each progression problem have been generated using Serpent 2, a continuous-energy Monte Carlo reactor physics burnup calculation code.Using the neutron multiplication eigenvalue ke as a metric, MPACT’s performance is assessed on each of the progression problems. Initial results showed that MPACT’s multi-group cross section libraries, originally developed for pressurized water reactor problems, were not sufficient to accurately solve Magnox problems. MPACT’s improved performance on the progression problems is demonstrated using this new optimized cross section library.

Luciano, Nicholas↗
DOE OSTI2020

BASS: An R Package for Fitting and Performing Sensitivity Analysis of Bayesian Adaptive Spline Surfaces

In this work, we present the R package BASS as a tool for nonparametric regression. The primary focus of the package is fitting fully Bayesian adaptive spline surface (BASS) models and performing global sensitivity analyses of these models. The BASS framework is similar to that of Bayesian multivariate adaptive regression splines (BMARS) from Denison, Mallick, and Smith (1998), but with many added features. The software is built to efficiently handle significant amounts of data with many continuous or categorical predictors and with functional response. Under our Bayesian framework, most priors are automatic but these can be modified by the user to focus on parsimony and the avoidance of overfitting. If directed to do so, the software uses parallel tempering to improve the reversible jump Markov chain Monte Carlo (RJMCMC) methods used to perform inference. We discuss the implementation of these features and present the performance of BASS in a number of analyses of simulated and real data.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

Utilizing ink composition to tune bulk-electrode gas transport, performance, and operational robustness for a Fe-N-C catalyst in polymer electrolyte fuel cell

With lower site density and turnover frequency, platinum group metal (PGM)-free catalysts based electrodes are often greater than 50 mu m thick in order to increase performance across the fuel cell operating range. Consequently, PGM-free electrodes have an additional bulk electrode transport resistance beyond the local or aggregate level transport in thin platinum-based electrodes. In parallel to the development of more active and durable PGM-free catalysts, advancements in understanding the interplay between PGM-free electrode fabrication, bulk-electrode transport, proton conductivity and performance are needed. Here, the relationship between ionic and gas phase transport through the electrode thickness is modified by adjusting electrocatalyst and ionomer flocculation/interaction at the ink level. The influence of the ink composition (water/n-propanol content) is examined via various in-situ electrochemical and ex-situ characterization techniques and the resulting electrode structure/performance relationship contrasted with electrode performance robustness across a range of relative humidity (RH). For the electrocatalyst examined here, a water-rich (82 wt% H2O) ink formulation was favorable for operation at high RH due to improved molecular diffusion through larger electrode pores. In contrast, the improved interactions between ionomer and electrocatalyst enabled a more robust electrode and higher performance during low RH operation for the 50 wt% H2O content ink.

08 HYDROGEN↗
DOE OSTI2024

Computational Analysis of Hydraulic Efficiency of Michigan DOT Cover C

Drainage structures are used to capture stormwater runoff in streets and highways in urban environments. These drainage structures, which typically consist of catch basins with grates, inlets, or combination grates/inlets, collect stormwater runoff and discharge through buried conveyance systems. They are strategically placed for public safety in curb and gutter systems to provide efficient drainage of water from roadways and thus reduce the risk of hydroplaning. The performance of drainage structures is measured in terms of hydraulic efficiency, which is defined as the percentage of flow captured by the basin as compared to the total flow drainage to the structure. Understanding of the performance of these drainage structures helps designers properly space inlets to promote an economic design that ensures the safety of the traveling public. The current design methodology used to determine drainage structure follows guidelines established in the current Michigan Department of Transportation (MDOT) Drainage Manual (2006). The guidelines in the MDOT Drainage Manual were modeled after the Federal Highway Administration’s (FHWA) Hydraulic Engineering Circular 22 “Urban Drainage Design” (HEC-22). HEC-22 includes empirically derived equations to calculate the interception capacity of drainage structures for several commonly used grate configurations, such as the parallel bar, curved vane and tilt bar grates, which are based on a research study performed by Burgi et al. in the 1970s. MDOT uses several drainage structures to capture runoff that are detailed as Standard Plans. Many of these drainage structures utilize sinusoidal type grates that are not described in HEC-22. Physical modeling of these structures has been limited, posing the need to have them analyzed to verify their capture efficiency. Current MDOT practice is to assume a similar sized reticuline grate, as described in HEC-22, for capture efficiencies. Until recently, evaluating the hydraulic performance of drainage structures was limited to physical modeling in a hydraulics laboratory. With advances in engineering software and computing power, computational fluid dynamics (CFD) modeling has become a more cost-effective alternative. The Federal Highway Administration (FHWA) provides states the option to evaluate their drainage structures using CFD through the Transportation Pooled Fund Program. This study, “Computational Analysis of Hydraulic Efficiency of Michigan DOT Cover C,” was carried out using the pooled fund. MDOT’s Cover C was chosen as the first test candidate, given its similar sinusoidal pattern to other MDOT grates, but it is typically used for high-volume, higher speed applications. A similar version, Cover CX, is used on interstate highways but does not have traverse bars for bicycle safety. Additional grates may be considered for evaluation in the future.

42 ENGINEERING↗
DOE OSTI2020

Improving Performance of M-to-N Processing and Data Redistribution in In Transit Analysis and Visualization

In an in transit setting, a parallel data producer, such as a numerical simulation, runs on one set of ranks M, while a data consumer, such as a parallel visualization application, runs on a different set of ranks N. One of the central challenges in this in transit setting is to determine the mapping of data from the set of M producer ranks to the set of N consumer ranks. This is a challenging problem for several reasons, such as the producer and consumer codes potentially having different scaling characteristics and different data models. The resulting mapping from M to N ranks can have a significant impact on aggregate application performance. In this work, we present an approach for performing this M-to-N mapping in a way that has broad applicability across a diversity of data producer and consumer applications. We evaluate its design and performance with a study that runs at high concurrency on a modern HPC platform. By leveraging design characteristics, which facilitate an “intelligent” mapping from M-to-N, we observe significant performance gains are possible in terms of several different metrics, including time-to-solution and amount of data moved.

Loring, Burlen↗
DOE OSTI2020

Improving Performance of M-to-N Processing and Data Redistribution in In Transit Analysis and Visualization

In an in transit setting, a parallel data producer, such as a numerical simulation, runs on one set of ranks M, while a data consumer, such as a parallel visualization application, runs on a different set of ranks N: One of the central challenges in this in transit setting is to determine the mapping of data from the set of M producer ranks to the set of N consumer ranks. This is a challenging problem for several reasons, such as the producer and consumer codes potentially having different scaling characteristics and different data models. The resulting mapping from M to N ranks can have a significant impact on aggregate application performance. In this work, we present an approach for performing this M-to-N mapping in a way that has broad applicability across a diversity of data producer and consumer applications. We evaluate its design and performance with a study that runs at high concurrency on a modern HPC platform. By leveraging design characteristics, which facilitate an ''intelligent'' mapping from M-to-N, we observe significant performance gains are possible in terms of several different metrics, including time-to-solution and amount of data moved.

Loring, Burlen↗
DOE OSTI2023

Multigrid Reduction in Time for Chaotic and Hyperbolic Problems (Final Report)

The coming massive parallelism of exascale computing presents a pressing challenge for the many DOE simulations of time-dependent partial differential equations (PDEs), which typically use traditional sequential time stepping methods. Since this traditional approach is inherently serial, it presents a sequential bottleneck when moving to exascale computing, because future performance gains will come through greater concurrency, not faster clock speeds. Thus, the goal of this work is to research parallelism in time, i.e., methods that compute multiple time values simultaneously, not sequentially. The focus will be on hyperbolic and chaotic problems of interest to DOE, with the goal of enabling scalable simulations of time-dependent hyperbolic and chaotic problems on future architectures. The chosen methodology for solving these problems parallel-in-time is multigrid, because multigrid (when it works) is a powerful, optimal, and scalable solver for discretized PDEs. Multigrid is already commonly used in many DOE simulations for scalably and optimally solving space-only PDE problems. The areas of hyperbolic and chaotic problems are chosen because of their relevance to problems of programmatic interest to DOE. However, these problems are also well-known to be difficult for parallelin-time methods, with the most common method, parareal, diverging in many cases. The current state-of-the-art for parallel-in-time at LLNL is the multigrid reduction in time (MGRIT) XBraid package, which also struggles for such problems, while still showing some improvement over parareal. In summary, new methods are needed for an efficient parallel-in-time scheme for hyperbolic and chaotic problems, and this work shall research promising new multigrid methods in this area. In particular, this work shall continue researching the directions from the current collaboration with Dr. Falgout, which are laid out in the work Toward Parallel in Time for Chaotic Dynamical Systems and showed the first known results of a parallel-in-time speedup for a chaotic problem. This work outlines two key improvements to XBraid for chaotic problems, the so-called “theta” and “delta-correction” methods. Here, these two improvements will be implemented in a high-performance but general way in XBraid and explored for more complicated problems. We will additionally research, as time allows, improvements to these techniques, as well as multigrid relaxation techniques based on Least Squares Shadowing (LSS by Wang) and a nonintrusive block tridiagonal solver based on MGRIT, called TriMGRIT.

97 MATHEMATICS AND COMPUTING↗