Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Data parallelism”

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 1,153 records · Page 64
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Operational experience and R&D results using the Google Cloud for High-Energy Physics in the ATLAS experiment

The ATLAS experiment at CERN relies on a Worldwide Distributed Computing Grid infrastructure to support its physics program at the Large Hadron Collider. ATLAS has integrated cloud computing resources to complement its Grid infrastructure and conducted an R&D program on Google Cloud Platform. These initiatives leverage key features of commercial cloud providers: lightweight configuration and operation, elasticity and availability of diverse infrastructures. Here this paper examines the seamless integration of cloud computing services as a conventional Grid site within the ATLAS workflow management and data management systems, while also offering new setups for interactive, parallel analysis. It underscores pivotal results that enhance the on-site computing model and outlines several R&D projects that have benefited from large-scale, elastic resource provisioning models. Furthermore, this study discusses the impact of cloud-enabled R&D projects in three domains: accelerators and AI/ML, ARM CPUs and columnar data analysis techniques.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
NASA NTRS2016

Implementing Connected Component Labeling as a User Defined Operator for SciDB

We have implemented a flexible User Defined Operator (UDO) for labeling connected components of a binary mask expressed as an array in SciDB, a parallel distributed database management system based on the array data model. This UDO is able to process very large multidimensional arrays by exploiting SciDB's memory management mechanism that efficiently manipulates arrays whose memory requirements far exceed available physical memory. The UDO takes as primary inputs a binary mask array and a binary stencil array that specifies the connectivity of a given cell to its neighbors. The UDO returns an array of the same shape as the input mask array with each foreground cell containing the label of the component it belongs to. By default, dimensions are treated as non-periodic, but the UDO also accepts optional input parameters to specify periodicity in any of the array dimensions. The UDO requires four stages to completely label connected components. In the first stage, labels are computed for each subarray or chunk of the mask array in parallel across SciDB instances using the weighted quick union (WQU) with half-path compression algorithm. In the second stage, labels around chunk boundaries from the first stage are stored in a temporary SciDB array that is then replicated across all SciDB instances. Equivalences are resolved by again applying the WQU algorithm to these boundary labels. In the third stage, relabeling is done for each chunk using the resolved equivalences. In the fourth stage, the resolved labels, which so far are "flattened" coordinates of the original binary mask array, are renamed with sequential integers for legibility. The UDO is demonstrated on a 3-D mask of O(1011) elements, with O(108) foreground cells and O(106) connected components. The operator completes in 19 minutes using 84 SciDB instances.

UDO↗
DOE OSTI2023

Invited Paper: Benchmarking and Optimizing Data Movement on Emerging Heterogeneous Architectures

As supercomputers evolve, nodes are continually increasing in complexity. As a result, each generation of parallel systems brings new performance challenges. For instance, on recent systems inter-node communication has outperformed inter-socket, resulting in poor performance of many node-aware communication optimizations. Communication optimizations are critical for the performance and scalability of parallel applications, but are dependent on the parallel architecture, which varies significantly among recent generations of supercomputers. Furthermore, this paper investigates the performance of various paths of data movement on recent generations of systems, and analyzes the increased complexity of communication, particularly on recent heterogeneous systems. The paper also introduces MPI Advance, a communication library that enables optimizations to be created based on benchmark analysis of each emerging system.

benchmarking↗
DOE OSTI2022

A Hybrid Data-Driven and Model-Based Anomaly Detection Scheme for DER Operation: Preprint

This paper proposes a hybrid data and model-based anomaly detection for securing the operation of distributed energy resources (DERs) in distribution grids. Data-driven autoencoders (AE) are set up at the edge level by taking local DER data and detect anomalous operations by leveraging the reconstruction ability. In parallel, model-based state estimation (SE) is running at the system level by taking system models and measurements, the anomalies are identified by analyzing the measurements residual. The hybrid scheme preserves the benefits of both data-driven and model-based analysis and thus improves the robustness and accuracy of anomaly detection. It can be established by getting full use of the existing infrastructures in distribution grids. Numerical tests on a realistic distribution feeder in Southern California highlight the effectiveness as well as benefits of the proposed scheme.

anomaly detection↗
DOE OSTI2025

Data for Kim et al., "Variations in the optical and molecular composition of dissolved organic matter exported from coastal wetlands"

Knowledge about sources and composition of marsh-derived dissolved organic matter (DOM) is critical for understanding the role of marshes in coastal biogeochemical cycling and the fate of marsh-derived DOM in the ocean. To investigate tidal variability in composition of marsh-derived DOM, Kim et al. examined the optical and molecular characteristics of hourly surface water samples at three tidal creeks in the Chesapeake Bay. Groundwater samples along the terrestrial landscape gradient as well as estuarine water from the adjacent estuary at each site were also collected to help resolve sources of surface water DOM. Samples were collected in summer 2024 at three sites – SWH: Sweet Hall Marsh, GCW: Kirkpatrick Marsh, and GWI: Goodwin Islands – which are part of synoptic sites in the Chesapeake Bay region of the COMPASS-FME (Coastal Observations, Mechanisms, and Predictions Across Systems and Scales - Field, Measurements, and Experiments) project. Surface water samples were collected hourly over a 48-hour period at each site. Groundwater and estuarine water samples were collected once. This dataset includes- Surface water depth and salinity- Dissolved organic carbon (DOC) and total dissolved nitrogen (TDN) concentrations- Optical indices and relative composition of parallel factor analysis (PARAFAC) components- High resolution mass spectrometry data.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS1988

Modal identification using single-mode projection filters and comparison with ERA and MLE results

The Single-Mode Projection Filter (SPF) is a newly developed algorithm for eigensystem parameter identification from both analytical results and test data. The SPF is formulated with a single mode only and practical for parallel processing implementation. Explicit formulations of SPF are derived for the multi-input multi-output (MIMO) system by using the orthogonal matrices of the controllability and observability matrices in the general sense. The modal parameters of SPF are initially obtained from an analytical model in modal space. The experimental data are then processed through SPF to update its modal parameters and to minimize a cost function defined by the norm of an error matrix. The updated modal parameters represent the characteristics of the test data. A two-dimensional global minimum optimization algorithm is developed and applied for the filter update by using the interval analysis method. The SPF is developed based on a single-mode subsystem and identifies only one modal frequency and one modal damping within a specified region. For an n-modes structure, n SPF can be implemented for parallel processing to reduce the computational burden. The SPF is applied to analyze the simulated data for the MAST beam structure. The estimated modal parameters are comparable to those from the Eigensystem Realization Algorithm (ERA) and repeated modal frequencies are identified. The modal analysis of the Spacecraft Control Laboratory Experiment (SCOLE) data is also performed by using the ERA and the Maximum Likelihood Estimate (MLE). The result shows that the first five modal frequencies are very close from ERA and MLE. However, there are slight disparities in the damping rates and the computational burdens are quite different among these two algorithms.

Huang, Jen-Kuang↗
NASA NTRS1990

General-Purpose Data-Formatting Input/Output System

Multiplexable input/output (I/O) system developed as interface between host computer and real-time, fixed-base simulator cockpit. Unit designed to be general-purpose interface and operated through virtually any 8- or 16-bit, transistor/transistor-logic-level, parallel-I/O port with or without handshaking. Accepts byte-coded data in form of data blocks from host computer and routes data to various system I/O modules like discrete outputs, lamp drivers, and seven-segment-display drivers. Formats and routes data from input modules back to host computer. Other modules designed to drive alphanumeric displays and communications interfaces and to serve as analog-to-digital and digital-to-analog converter modules.

Busquets, Anthony M.↗
NASA NTRS1989

Efficient ICCG on a shared memory multiprocessor

Different approaches are discussed for exploiting parallelism in the ICCG (Incomplete Cholesky Conjugate Gradient) method for solving large sparse symmetric positive definite systems of equations on a shared memory parallel computer. Techniques for efficiently solving triangular systems and computing sparse matrix-vector products are explored. Three methods for scheduling the tasks in solving triangular systems are implemented on the Sequent Balance 21000. Sample problems that are representative of a large class of problems solved using iterative methods are used. We show that a static analysis to determine data dependences in the triangular solve can greatly improve its parallel efficiency. We also show that ignoring symmetry and storing the whole matrix can reduce solution time substantially.

Hammond, Steven W.↗
NASA NTRS1994

Low-freqency instabilities and the resulting velocity distributions of pickup ions at comet Halley

The interaction between the solar wind and newborn cometary ions is studied using a new analytical theory as well as one- and two-dimensional hybrid simulations. Using the observed parameters upstream of the comet Halley, a detailed study of wave excitation and the resulting particle distributions is presentd. Linear theory as well as simulations show that a variety of modes such as the fast magnetosonic mode, high frequency whistlers and obliquely propagating Alfven ion cyclotron waves can be excited. However, parallel propagating waves are found to be dominant in the wave spectrum and to control the scattering of the pickup ions. Several features of the observed distributions of pickup protons are explained. In particular, it is shown that the observed asymmetric pitch distribution for the pickup protons is due to the small saturation amplitude of the waves for the given parameters. Water group associated waves can lead to energy diffusion and further pitch angle scattering of protons. This effect is most likely to be important in the vicinity of the bow shock of comet Halley where the density of water group ions becomes comparable to that of protons. It is shown that the observed increase in the radius of the proton velocity shell just outside the bow shock can be due to water group waves. The nearby isotropic proton pitch angle distribution observed by Neugebauer et al. (1989) just outside the bow shock may, however, be related to the presence of a rotational discontinuity which has been identified in the magnetic field data. Just outside the bow shock, simulations show that parallel propagating water group waves can steepen with attached whistler wave packets. The steepening process at parallel propagation is a transient effect, in an important contrast to the case of steepening at oblique angles. The smaller beam densitites at comet Halley appears to be the main reason not only why waves at comet Halley have smaller amplitudes but also why oblique, steepening magnetosonic waves have not been detected at comet Halley, whereas they have been seen at comet Giacobini-Zinner. Halley, whereas they have been seen at comet Giacobini-Zinner.

Karimabadi, H.↗
DOE OSTI2024

Airborne LiDAR to Improve Canopy Fuels Mapping for Wildfire Modeling

Increasing conflict between wildfire and the built environment has increased the need for more up-to-date and finer resolution canopy fuels data to improve wildfire modeling and associated risk forecasts. The US Forest Service and US Department of the Interior’s LANDFIRE product, which provides 30-m resolution canopy fuels data for the entire US, is one of the most widely used sources of fuels data. However, the last complete mapping effort for LANDFIRE is based on 2016 conditions, and subsequent updates reflect disturbances 1-2 years behind the release year. Airborne systems equipped with Light Detection and Ranging (LiDAR) sensors can be deployed to actively sense canopy structure and estimate canopy fuels data (cover, height, base height, bulk density) at finer resolutions. Canopy base height (CBH) and canopy bulk density (CBD) are difficult to measure both in the field and in LiDAR point clouds. Still, they are important for accurately modeling crown fires, which are often intense and difficult to contain. Additionally, point cloud datasets are large, and calculations require efficient utilization of computational resources. To address these challenges, we are working on an approach that uses openly available National Ecological Observatory Network (NEON) airborne LiDAR data, with calculations processed in the R programming language and parallelized through the lidR package. CBH and CBD are often derived from tree height, diameter at breast height, and species-specific allometries using the Fire and Fuels Extension of the Forest Vegetation Simulator (FFE-FVS). We aim to test if airborne LiDAR can estimate CBH and CBD without the use of empirical equations. Reliable estimates of canopy fuels data directly from airborne LiDAR could streamline quick, fine-resolution updates for use in wildfire behavior models.

54 ENVIRONMENTAL SCIENCES↗
NASA NTRS2006

The Use of Dynamic Visual Acuity as a Functional Test of Gaze Stabilization Following Space Flight

After prolonged exposure to a given gravitational environment the transition to another is accompanied by adaptations in the sensorimotor subsystems, including the vestibular system. Variation in the adaptation time course of these subsystems, and the functional redundancies that exist between them make it difficult to accurately assess the functional capacity and physical limitations of astro/cosmonauts using tests on individual subsystems. While isolated tests of subsystem performance may be the only means to address where interventions are required, direct measures of performance may be more suitable for assessing the operational consequences of incomplete adaptation to changes in the gravitational environment. A test of dynamic visual acuity (DVA) is currently being used in the JSC Neurosciences Laboratory as part of a series of measures to assess the efficacy of a countermeasure to mitigate postflight locomotor dysfunction. In the current protocol, subjects visual acuity is determined using Landolt ring optotypes presented sequentially on a computer display. Visual acuity assessments are made both while standing and while walking at 1.8 m/s on a motorized treadmill. The use of a psychophysical threshold detection algorithm reduces the required number of optotype presentations and the results can be presented immediately after the test. The difference between the walking and standing acuity measures provides a metric of the change in the subject s ability to maintain gaze fixation on the visual target while walking. This functional consequence is observable regardless of the underlying subsystem most responsible for the change. Data from 15 cosmo/astronauts have been collected following long-duration (approx. 6 months) stays in space using a visual target viewing distance of 4.0 meters. An investigation of the group mean shows a change in DVA soon after the flight that asymptotes back to baseline approximately one week following their return to earth. The performance of some subjects nicely parallels the stereotypical recovery curve observed in the group mean data. Others show dramatic changes in DVA from one test day to another. These changes may be indicative of a re-adaptation process that is not characterized by a steady improvement with the passage of time, but is instead a dynamic search for appropriate coordinative strategy to achieve the desired gaze stabilization goal. Ground-based data have been collected in our lab using DVA with one of the goals being to improve the DVA test itself. In one of these studies, the DVA test was repeated using a visual target viewing distance of 0.5 meters. While walking, the relative contributions of the otoliths and semi-circular canals that are required to stabilize gaze are affected by visual target viewing distance. It may be possible to exploit this using the current treadmill DVA test to differentially assess changes in these vestibular subsystems. The postflight DVA evaluations currently used have been augmented to include the near target version of the test. Preliminary results from these assessments, as well as the results from the ground-based tests will also be reported. DVA provides a direct measure of a subject's ability to see clearly in the presence of self-motion. The use of the current tests for providing a functionally relevant metric is evident. However, it is possible to expand the scope of DVA testing to include scenarios other than walking. A facility for measuring DVA in the presence of passive movements is being created. Using a mechanized platform to provide the perturbation, it should be possible to simulate aircraft and automobile vibration profiles. Used in conjunction with the far and near visual displays this facility should be able to assess a subject s ability to clearly see distant objects as well as those that appear on the dashboard or instrument control panel during functionally relevant situations.

Peters, B. T.↗
NASA NTRS2020

Single-Photon Counting Detector Scalability for High Photon Efficiency Optical Communications Links

For high photon-efficiency deep space or low power optical communications links, such as the Orion Artemis-2 Optical Communications System (O2O) project, the received optical signal is attenuated to the extent that single- photon detectors are required. For direct-detection receivers operating at 1.55 µm wavelength, single-photon detectors including Geiger-mode InGaAs avalanche photon diodes (APDs), and in particular superconducting nanowire single-photon detectors (SNSPDs) offer the highest sensitivity and fastest detection speeds. However, these photon detectors exhibit a recovery time between registered input pulses, effectively reducing the detection efficiency over the recovery interval, resulting in missed photon detections, reduced count rate, and ultimately limiting the achievable data rate. A method to overcome this limitation is to divide the received optical signal into multiple detectors in parallel. Here we analyze this approach for a receiver designed to receive a high photon efficiency serially concatenated pulse position modulation (SCPPM) input waveform. From measured count rate and efficiency data using commercial SNSPDs, we apply a model from which we determine the effective detection efficiency, or blocking loss, for different input signal rates. We analyze the scalability of adding detectors in parallel for different modulation orders and background levels to achieve desired data rates. Finally we show tradeoffs between the number of detectors and the required received optical power, useful for real link design considerations.

Vyhnalek, Brian E.↗
NASA NTRS1992

Modeling of supersonic combustor flows using parallel computing

While current 3D CFD codes and modeling techniques have been shown capable of furnishing engineering data for complex scramjet flowfields, the usefulness of such efforts is primarily limited by solutions' CPU time requirements, and secondarily by memory requirements. Attention is presently given to the use of parallel computing capabilities for engineering CFD tools for the analysis of supersonic reacting flows, and to an illustrative incompressible CFD problem using up to 16 iPSC/2 processors with single-domain decomposition.

Riggins, D.↗
NASA NTRS1999

A New Approach to Parallel Dynamic Partitioning for Adaptive Unstructured Meshes

Classical mesh partitioning algorithms were designed for rather static situations, and their straightforward application in a dynamical framework may lead to unsatisfactory results, e.g., excessive data migration among processors. Furthermore, special attention should be paid to their amenability to parallelization. In this paper, a novel parallel method for the dynamic partitioning of adaptive unstructured meshes is described. It is based on a linear representation of the mesh using self-avoiding walks.

Heber, Gerd↗
DOE OSTI2021

Multimodal Bayesian registration of noisy functions using Hamiltonian Monte Carlo

Functional data registration is a necessary processing step for many applications. The observed data can be inherently noisy, often due to measurement error or natural process uncertainty; which most functional alignment methods cannot handle. A pair of functions can also have multiple optimal alignment solutions, which is not addressed in current literature. In this paper, a flexible Bayesian approach to functional alignment is presented, which appropriately accounts for noise in the data without any pre-smoothing required. Additionally, by running parallel MCMC chains, the method can account for multiple optimal alignments via the multi-modal posterior distribution of the warping functions. To most efficiently sample the warping functions, the approach relies on a modification of the standard Hamiltonian Monte Carlo to be well-defined on the infinite-dimensional Hilbert space. In this work, this flexible Bayesian alignment method is applied to both simulated data and real data sets to show its efficiency in handling noisy functions and successfully accounting for multiple optimal alignments in the posterior; characterizing the uncertainty surrounding the warping functions.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Efficient Distributed Sequence Parallelism for Transformer-Based Image Segmentation

We introduce an efficient distributed sequence parallel approach for training transformer-based deep learning image segmentation models. The neural network models are comprised of a combination of a Vision Transformer encoder with a convolutional decoder to provide image segmentation mappings. The utility of the distributed sequence parallel approach is especially useful in cases where the tokenized embedding representation of image data are too large to fit into standard computing hardware memory. To demonstrate the performance and characteristics of our models trained in sequence parallel fashion compared to standard models, we evaluate our approach using a 3D MRI brain tumor segmentation dataset. We show that training with a sequence parallel approach can match standard sequential model training in terms of convergence. Furthermore, we show that our sequence parallel approach has the capability to support training of models that would not be possible on standard computing resources.

Lyngaas, Isaac↗
NASA NTRS1992

The role of HiPPI switches in mass storage systems: A five year prospective

New standards are evolving which provide the foundation for novel multi-gigabit per second data communication structures. The lowest layer protocols are so generalized that they encourage a wide range of application. Specifically, the ANSI High Performance Parallel Interface (HiPPI) is being applied to computer peripheral attachment as well as general data communication networks. This paper introduces the HiPPI standards suite and technology products which incorporate the standards. The use of simple HiPPI crosspoint switches to build potentially complex extended 'fabrics' is discussed in detail. Several near term applications of the HiPPI technology are briefly described with additional attention to storage systems. Finally, some related standards are mentioned which may further expand the concepts above.

Gilbert, T. A.↗
NASA NTRS1991

The role of HiPPI switches in mass storage systems: A five year prospective

New standards are evolving which provide the foundation for multi-gigabit per second data communication structures. The lowest layer protocols are so generalized that they encourage a wide range of application. Specifically, the ANSI High Performance Parallel Interface (HiPPI) is being applied to computer peripheral attachment as well as general data communication networks. The HiPPI Standards suite and technology products which incorporate the standards are introduced. The use of simple HiPPI crosspoint switches to build potentially complex extended 'fabrics' is discussed in detail. Several near term applications of the HiPPI technology are briefly described with additional attention to storage systems. Finally, some related standards are mentioned which may further expand the concepts above.

Gilbert, T. A.↗