Search NASA⌕ Search

SEARCH · Search NASA

Results for “data parallel”

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 1,747 records · Page 97
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Modeling performance of data collection systems for high-energy physics

Exponential increases in scientific experimental data are outpacing silicon technology progress, necessitating heterogeneous computing systems—particularly those utilizing machine learning (ML)—to meet future scientific computing demands. The growing importance and complexity of heterogeneous computing systems require systematic modeling to understand and predict the effective roles for ML. We present a model that addresses this need by framing the key aspects of data collection pipelines and constraints and combining them with the important vectors of technology that shape alternatives, computing metrics that allow complex alternatives to be compared. For instance, a data collection pipeline may be characterized by parameters such as sensor sampling rates and the overall relevancy of retrieved samples. Alternatives to this pipeline are enabled by development vectors including ML, parallelization, advancing CMOS, and neuromorphic computing. By calculating metrics for each alternative such as overall F1 score, power, hardware cost, and energy expended per relevant sample, our model allows alternative data collection systems to be rigorously compared. We apply this model to the Compact Muon Solenoid experiment and its planned high luminosity-large hadron collider upgrade, evaluating novel technologies for the data acquisition system (DAQ), including ML-based filtering and parallelized software. The results demonstrate that improvements to early DAQ stages significantly reduce resources required later, with a power reduction of 60% and increased relevant data retrieval per unit power (from 0.065 to 0.31 samples/kJ). However, we predict that further advances will be required in order to meet overall power and cost constraints for the DAQ.

Olin-Ammentorp, Wilkie (ORCID:0000000224729862)↗
NASA NTRS1991

One GHz digitizer for space based laser altimeter

This is the final report for the research and development of the one GHz digitizer for space based laser altimeter. A feasibility model was designed, built, and tested. Only partial testing of essential functions of the digitizer was completed. Hybrid technology was incorporated which allows analog storage (memory) of the digitally sampled data. The actual sampling rate is 62.5 MHz, but executed in 16 parallel channels, to provide an effective sampling rate of one GHz. The average power consumption of the one GHz digitizer is not more than 1.5 Watts. A one GHz oscillator is incorporated for timing purposes. This signal is also made available externally for system timing. A software package was also developed for internal use (controls, commands, etc.) and for data communication with the host computer. The digitizer is equipped with an onboard microprocessor for this purpose.

Staples, Edward J.↗
DOE OSTI2015

Aberration‐Corrected Transmission Electron Microscopy and In Situ XAFS Structural Characterization of Pt/γ‐Al 2 O 3 Nanoparticles

Abstract Aberration‐corrected (AC) STEM, AC TEM and in situ X‐ray absorption fine structure spectroscopy (XAFS) were used to characterize the Pt clusters present on a 0.35 wt % Pt on γ‐alumina support after reduction in hydrogen at 700 °C. STEM high‐angle annular dark field imaging shows that cluster formation takes place at temperatures up to approximately 350 °C, and this is followed by gradual growth in cluster size for heat treatments in hydrogen up to 700 °C. The STEM data show that after 700 °C reduction the Pt clusters are present in a narrow size distribution centered at 0.88 nm, and using a method involving a redistribution of the Pt atoms using a high electron dosage in the STEM, it is shown that the clusters are present in two‐dimensional morphology. This conclusion is verified using intensity line scans. The in situ extended X‐ray absorption fine structure data are in good agreement with these observations. High‐resolution AC–TEM, which uses a broad coherent electron beam, and can thus offer advantages relative to STEM for structure determination of fine clusters, supported by image simulations of through‐focus series, were used to analyze the structures of Pt particles. The structures determined by using AC–TEM are consistent with STEM and EXAFS data in having a flat two‐dimensional morphology. Comparison of AC–STEM and AC TEM data for the same 700 °C reduced sample suggests that parallel‐beam TEM mode of imaging may be advantageous because of the less pronounced beam‐induced structural rearrangements that occur when imaging with a fine STEM probe.

Sinkler, Wharton↗
DOE OSTI2015

TECA: Petascale pattern recognition for climate science

Climate Change is one of the most pressing challenges facing humanity in the 21st century. Climate simulations provide us with a unique opportunity to examine effects of anthropogenic emissions. Highresolution climate simulations produce “Big Data”: contemporary climate archives are ≈ 5PB in size and we expect future archives to measure on the order of Exa-Bytes. In this work, we present the successful application of TECA (Toolkit for Extreme Climate Analysis) framework, for extracting extreme weather patterns such as Tropical Cyclones, Atmospheric Rivers and Extra-Tropical Cyclones from TB-sized simulation datasets. TECA has been run at full-scale on Cray XE6 and IBM BG/Q systems, and has reduced the runtime for pattern detection tasks from years to hours. TECA has been utilized to evaluate the performance of various computational models in reproducing the statistics of extreme weather events, and for characterizing the change in frequency of storm systems in the future.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Characterizing the performance of node-aware strategies for irregular point-to-point communication on heterogeneous architectures

Supercomputer architectures are trending toward higher computational throughput due to the inclusion of heterogeneous compute nodes. These multi-GPU nodes increase on-node computational efficiency, while also increasing the amount of data to be communicated and the number of potential data flow paths. In this work, we characterize the performance of irregular point-to-point communication with MPI on heterogeneous compute environments through performance modeling, demonstrating the limitations of standard communication strategies for both device-aware and staging-through-host communication techniques. Presented models suggest staging communicated data through host processes then using node-aware communication strategies for high inter-node message counts. Notably, the models also predict that node-aware communication utilizing all available CPU cores to communicate inter-node data leads to the most performant strategy when communicating with a high number of nodes. Furthermore, model validation is provided via a case study of irregular point-to-point communication patterns in distributed sparse matrix–vector products. Importantly, we include a discussion on the implications model predictions have on communication strategy design for emerging supercomputer architectures.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

Laser Ablation-Aerosol Mass Spectrometry-Chemical Ionization Mass Spectrometry for Ambient Surface Imaging

Mass spectrometry imaging is becoming an increasingly common analytical technique due to its ability to provide spatially resolved chemical information. In this paper, we report a novel imaging approach combining laser ablation with two mass spectrometric techniques, aerosol mass spectrometry and chemical ionization mass spectrometry, separately and in parallel. Both mass spectrometric methods provide the fast response, rapid data acquisition, low detection limits, and high-resolution peak separation desirable for imaging complex samples. Additionally, the two techniques provide complementary information with aerosol mass spectrometry providing near universal detection of all aerosol molecules and chemical ionization mass spectrometry with a heated inlet providing molecular-level detail of both gases and aerosols. The two techniques operate with atmospheric pressure interfaces and require no matrix addition for ionization, allowing for samples to be investigated in their native state under ambient pressure conditions. We demonstrate the ability of laser ablation-aerosol mass spectrometry-chemical ionization mass spectrometry (LA-AMS-CIMS) to create 2D images of both standard compounds and complex mixtures. Finally, the results suggest that LA-AMS-CIMS, particularly when combined with advanced data analysis methods, could have broad applications in mass spectrometry imaging applications.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2019

Polarized Soft X-ray Scattering Reveals Chain Orientation within Nanoscale Polymer Domains

Fully conjugated donor-acceptor block copolymers can serve as the active layer in organic photovoltaics and other organic electronic devices. Additionally, self-assembly into periodic domains and crystallization of the constituent blocks are crucial to enable control of the multiscale structure and consequently electronic properties. Resonant soft X-ray scattering (RSoXS) is an invaluable tool to characterize such materials, where tuning the X-ray energy and polarization reveals molecular orientation and domain spacing. In this work, anisotropic soft X-ray scattering data reveal the type and degree of orientation within conjugated block copolymers composed of poly(3-hexylthiophene-2,5-diyl) and various push-pull copolymers. The maximum anisotropy is observed at the scattering vector corresponding to the spacing between block copolymer domains, which scales with the end-to-end distance of the blocks, as expected. Furthermore, the anisotropy in RSoXS data reveals that the crystalline P3HT blocks orient, on average, parallel or nearly parallel to the block copolymer interface; the average tilt angle between P3HT chains and domain interfaces can be extracted from the dependence of the anisotropy with polar angle. We interpret this angle to correspond to the average tilt of rings in the unit cell, potentially within a chiral mesostructure. Results are corroborated with scattering calculations based on simple model structures.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2015

Collective I/O Tuning Using Analytical and Machine-Learning Models

The ever larger demand of scientific applications for computation and data is currently driving a continuous increase in scale of parallel computers. The inherent complexity of scaling up a computing systems in terms of both hardware and software stack exposes an increasing number of factors impacting the performance and complicating the process of optimization. In particular, the optimization of parallel I/O has become increasingly challenging due to increasing storage hierarchy and well known performance variability of shared storage systems. This paper focuses on model-based autotuning of the two-phase collective I/O algorithm from a popular MPI distribution on the Blue Gene/Q architecture. We propose a novel hybrid model, constructed as a composition of analytical models for communication and storage operations and black-box models for the performance of the individual operations. We perform an in-depth study of the complexity involved in performance modeling including architecture, software stack and noise. In particular we address this challenges of modeling the performance of shared storage systems by building a benchmark that helps synthesizing factors such as topology, file caching, and noise. The experimental results show that the hybrid approach produces significantly better results than state-of-the-art machine learning approaches and shows a higher robustness to noise, at the cost of a higher modeling complexity

I/O performance modeling↗
DOE OSTI2017

Exacution: Enhancing Scientific Data Management for Exascale

As we continue toward exascale, scientific data volume is continuing to scale and becoming more burdensome to manage. In this paper, we lay out opportunities to enhance state of the art data management techniques. We emphasize well-principled data compression, and using it to achieve progressive refinement. This can both accelerate I/O and afford the user increased flexibility when she interacts with the data. The formulation naturally maps onto enabling partitioning of the progressively improving-quality representations of a data quantity into different media-type destinations, to keep the highest priority information as close as possible to the computation, and take advantage of deepening memory/storage hierarchies in ways not previously possible. Careful monitoring is requisite to our vision, not only to verify that compression has not eliminated salient features in the data, but also to better understand the performance of massively parallel scientific applications. Increased mathematical rigor would be ideal,to help bring compression on a better-understood theoretical footing, closer to the relevant scientific theory, more aware of constraints imposed by the science, and more tightly error-controlled. Throughout, we highlight pathfinding research we have begun exploring related these topics, and comment toward future work that will be needed.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

Distributed Training for High Resolution Images: A Domain and Spatial Decomposition Approach

In this work we developed two Pytorch libraries using the PyTorch RPC interface for distributed deep learning approaches on high resolution images. The spatial decomposition library allows for distributed training on very large images, which otherwise wouldn’t be possible on a single GPU. The domain parallelism library allows for distributed training across multiple domain unlabeled data, by leveraging the domain separation architecture. Both of those libraries where tested on the Summit supercomputer at Oak Ridge National Laboratory at a moderate scale.

Tsaris, Aristeidis (aris)↗
DOE OSTI2018

Python-based in situ analysis and visualization

This work focuses on enabling the use of Python-based methods for the purpose of performing in situ analysis and visualization. This approach facilitates access to and use of a rapidly growing collection of Python-based, third-party libraries for analysis and visualization, as well as lowering the barrier to entry for user-written Python analysis codes. Beginning with a simulation code that is instrumented to use the SENSEI in situ interface, we present how to couple it with a Python-based data consumer, which may be run in situ, and in parallel at the same concurrency as the simulation. We present two examples that demonstrate the new capability. One is an analysis of the reaction rate in a proxy simulation of a chemical reaction on a 2D substrate, while the other is a coupling of an AMR simulation to Yt, a parallel visualization and analysis library written in Python. In the examples, both the simulation and Python in situ method run in parallel on a large-scale HPC platform.

Loring, Burlen↗
DOE OSTI2022

A New Model for Simulating the Imbibition of a Wetting-Phase Fluid in a Matrix-Fracture Dual Connectivity System

The imbibition experiment is an effective approach for measuring petrophysical properties of porous media, with many such experiments performed over the past decade. Quite some empirical, analytical, and numerical models have been developed to simulate spontaneous imbibition of the wetting phase fluid into porous media, but limitations still exist. In previous studies, the imbibition process has been considered to give a piston-like displacement or the porous medium modeled as multiply-sized pores linked with bonds; both approaches fail to yield comprehensive results due to their neglect of the presence of irregular fractures or nonuniform flow paths through the matrix. By building a numerical model for simulating laboratory-scale experimental data, we performed imbibition tests on several fractured Barnett Shale samples having fractures either parallel ( P ) or transverse ( T ) to the bedding plane and used MATLAB to build a new numerical model by combining the imbibition process in fractures and the matrix using concepts from percolation theory. The experimental data show that the rocks with P -direction fractures have a more steady increase of imbibition rates than the case of T -direction one. As the shale matrix with low pore connectivity hampers the upward water movement, the imbibition rate of shales with T -direction fractures will decrease suddenly after the bottom layer in contact with water is saturated during the initial period. This wetting phase movement (WPM) model can simulate 3D porous media with 2D fractures. The rate of imbibition by fractured porous media is associated with physical parameters such as porosity and fracture distribution (e.g., the number and angle of fractures). Using Monte Carlo methods, we examined fracture parameters and predicted elapsed time and cumulative water imbibition, for the Barnett Shale samples. The results show that the rate of imbibed water mass is sensitive to the number of fractures directly connected to water source, and the connectivity between two neighboring grid cells is a key parameter for the wetting-front progression. The findings of this study can help to better understand the imbibition process with multiple influencing processes and factors in fractured-matrix rocks. Although the experiments, data simulation, and prediction results are based only on Barnett Shale samples, the model is readily applicable to imbibition tests of other fractured rocks to show the spatial and temporal behavior during a dynamic imbibition process that are not easily captured experimentally.

11 NUCLEAR FUEL CYCLE AND FUEL MATERIALS↗
DOE OSTI2024

batteryNET v0.0.1

Code to organize, prepare, and segment multi-scan, larger-than-memory, volumetric dataset. Deep learning segmentation uses pytorch-lightning for code organization, logging and metrics, and multi-GPU parallelization of training and prediction. Also uses moani for volumetric data augmentation and model implementation (currently U-Net++). Training can be done slice-by-slice with 2D models trained on 2D patches either aligned with or perpendicular to the electrode plane, or with 3D models on 3D patches. Sparse labels can be used to minimize required hand-labeling -- less than 0.1% of the entire dataset was labeled. Also includes code for data preparation (alignment/cropping/patching).

Ushizima, Daniela↗
DOE OSTI2019

Camellia: A Rapid Development Framework for Finite Element Solvers

We report the discontinuous Petrov–Galerkin (DPG) methodology of Demkowicz and Gopalakrishnan guarantees the optimality of the finite element solution in a user-controllable energy norm, and provides several features supporting adaptive schemes. The approach provides stability automatically; there is no need for carefully derived numerical fluxes (as in DG schemes) or for mesh-dependent stabilization terms (as in stabilized methods). In this paper, we focus on features of Camellia that facilitate implementation of new DPG formulations; chief among these is a rich set of features in support of symbolic manipulation, which allow, e.g., bilinear formulations in the code to appear much as they would on paper. Many of these features are general in the sense that they can also be used in the implementation of other finite element formulations. In fact, because DPG’s requirements are essentially a superset of those of other finite element methods, Camellia provides built-in support for most common methods. We believe, however, that the combination of an essentially “hands-free” finite element methodology as found in DPG with the rapid development features of Camellia are particularly winsome, so we focus on use cases in this class. In addition to the symbolic manipulation features mentioned above, Camellia offers support for one-irregular adaptive meshes in 1D, 2D, 3D, and space-time. It provides a geometric multigrid preconditioner particularly suited for DPG problems, and supports distributed parallel execution using MPI. For its load balancing and distributed data structures, Camellia relies on packages from the Trilinos project, which simplifies interfacing with other computational science packages. Camellia also allows loading of standard mesh formats through an interface with the MOAB package. Camellia includes support for static condensation to eliminate element-interior degrees of freedom locally, usually resulting in substantial reduction of the cost of the global problem. Finally, we include a discussion of the variational formulations built into Camellia, with references to those formulations in the literature, as well as an MPI performance study.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2017

UPC++ Programmer’s Guide (v1.0 2017.9)

UPC++ is a C++11 library that provides Asynchronous Partitioned Global Address Space (APGAS) programming. It is designed for writing parallel programs that run efficiently and scale well on distributed-memory parallel computers. The APGAS model is single program, multiple-data (SPMD), with each separate thread of execution (referred to as a rank, a term borrowed from MPI) having access to local memory as it would in C++. However, APGAS also provides access to a global address space, which is allocated in shared segments that are distributed over the ranks. UPC++ provides numerous methods for accessing and using global memory. In UPC++, all operations that access remote memory are explicit, which encourages programmers to be aware of the cost of communication and data movement. Moreover, all remote-memory access operations are by default asynchronous, to enable programmers to write code that scales well even on hundreds of thousands of cores.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2018

UPC++ Programmer’s Guide, v1.0-2018.3.0

UPC++ is a C++11 library that provides Partitioned Global Address Space (PGAS) programming. It is designed for writing parallel programs that run efficiently and scale well on distributed-memory parallel computers. The PGAS model is single program, multiple-data (SPMD), with each separate thread of execution (referred to as a rank, a term borrowed from MPI) having access to local memory as it would in C++. However, PGAS also provides access to a global address space, which is allocated in shared segments that are distributed over the ranks. UPC++ provides numerous methods for accessing and using global memory. In UPC++, all operations that access remote memory are explicit, which encourages programmers to be aware of the cost of communication and data movement. Moreover, all remote-memory access operations are by default asynchronous, to enable programmers to write code that scales well even on hundreds of thousands of cores.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2019

UPC++ Programmer’s Guide (v1.0-2019.3.0)

UPC++ is a C++11 library that provides Partitioned Global Address Space (PGAS) programming. It is designed for writing parallel programs that run efficiently and scale well on distributed-memory parallel computers. The PGAS model is single program, multiple-data (SPMD), with each separate constituent process having access to local memory as it would in C++. However, PGAS also provides access to a global address space, which is allocated in shared segments that are distributed over the processes. UPC++ provides numerous methods for accessing and using global memory. In UPC++, all operations that access remote memory are explicit, which encourages programmers to be aware of the cost of communication and data movement. Moreover, all remote-memory access operations are by default asynchronous, to enable programmers to write code that scales well even on hundreds of thousands of cores.

97 MATHEMATICS AND COMPUTING↗