SEARCH · Search NASA
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.
SF-23-067 BoxKit is a library that provides building blocks to parallelize and scale data science, high performance computing, and machine learning applications for block-structured datasets. Spatial data from simulations and experiments can be accessed and managed using tools available in this library when working with more data analysis oriented packages like SciKit (https://github.com/scikit-learn/scikit-learn) and FlowNet (https://github.com/NVIDIA/flownet2-pytorch)
The SIERRA Toolkit is a collection of libraries to facilitate the development of parallel engineering analysis applications. These libraries supply basic core services that an engineering application may need such as a parallel distributed and dynamic mesh database (for unstructured meshes), mechanics algorithm support (parallel infrastructure only), interfaces to parallel solvers, parallel mesh and data I/O, and various utilities (timers, diagnostic tools, etc.)
When parallel algorithms for simulation were introduced in the 1970s, their development and use interested only experts in parallel computation. This circumstance changed as multi-core processors became commonplace, putting a parallel computer into the hands of every modeler. A natural outcome is growing interest in parallel simulation among persons not intimately familiar with parallel computing. At the same time, parallel simulation tools continue to be developed with the implicit assumption that the modeler is knowledgeable about parallel programming. The unintended consequence is a rapidly growing number of users of parallel simulation tools that are unlikely to recognize when the interaction of race conditions, partitioning strategies, and simultaneous action in their simulation models make results non-reproducible, thereby calling into question the validity of conclusions drawn from the simulation data. Here, we illustrate the potential dangers of exposing parallel algorithms to users who are not experts in parallel computation with example models constructed using existing parallel simulation tools. By doing so, we hope to refocus tool developers on usability, even if this new focus incurs loss of some performance.
As parallel computing trends towards the exascale, scientific data produced by high-fidelity simulations are growing increasingly massive. For instance, a simulation on a three-dimensional spatial grid with 512 points per dimension that tracks 64 variables per grid point for 128 time steps yields 8 TB of data. By viewing the data as a dense five-way tensor, we can compute a Tucker decomposition to find inherent low-dimensional multilinear structure, achieving compression ratios of up to 10000 on real-world data sets with negligible loss in accuracy. So that we can operate on such massive data, we present the first-ever distributed-memory parallel implementation for the Tucker decomposition, whose key computations correspond to parallel linear algebra operations, albeit with nonstandard data layouts. Our approach specifies a data distribution for tensors that avoids any tensor data redistribution, either locally or in parallel. This software provides a method for compressing large-scale multiway data.
A method for processing data is provided. Data is identified by a computer system. The data is processed in parallel by the computer system using temporal transformations to form pieces of temporal data. The pieces of temporal data are placed by the computer system in an order as the pieces of temporal data are generated by the temporal transformations to form a sequence of temporal data. The order of the sequence is based on a priority of when the pieces of temporal data should be processed, enabling performing an action.
In recent years, the CNNs have achieved great successes in the image processing tasks, e.g., image recognition and object detection. Unfortunately, traditional CNN's classication is found to be easily misled by increasingly complex image features due to the usage of pooling operations, hence unable to preserve accurate position and pose information of the objects. To address this challenge, a novel neural network structure called Capsule Network has been proposed, which introduces equivariance through capsules to signicantly enhance the learning ability for image segmentation and object detection. Due to its requirement of performing a high volume of matrix operations, CapsNets have been generally accelerated on modern GPU platforms that provide highly optimized software library for common deep learning tasks. However, based on our performance characterization on modern GPUs, CapsNets exhibit low effciency due to the special program and execution features of their routing procedure, including massive unshareable intermediate variables and intensive syn- chronizations, which are very dicult to optimize at software level. To address these challenges, we propose a hybrid computing architecture design named PIM-CapsNet. It preserves GPU's on-chip computing capability for accelerating CNN types of layers in CapsNet, while pipelining with an off-chip in-memory acceleration solution that effectively tackles routing procedure's ineffciency by leveraging the processing-in-memory capability of today's 3D stacked memory. Using routing procedure's inherent parallellization feature, our design enables hierarchical improvements on CapsNet inference effciency through minimizing data movement and maximizing parallel processing in memory. Evaluation results demonstrate that our proposed design can achieve substantial improvement on both performance and energy savings for CapsNet inference, with almost zero accuracy loss. The results also suggest good performance scalability in optimizing the routing procedure with increasing network size.
Scientific applications at exascale generate and analyze massive amounts of data. A critical requirement of these applications is the capability to access and manage this data efficiently on exascale systems. As such, parallel I/O, the key technology enables moving data between compute nodes and storage, faces monumental challenges from new applications, memory, and storage architectures considered in the designs of exascale systems. As the storage hierarchy is expanding to include node-local persistent memory, burst buffers, etc., as well as disk-based storage, data movement among these layers must be efficient. Parallel I/O libraries of the future should be capable of handling file sizes of many terabytes and beyond. In this paper, we describe new capabilities we have developed in Hierarchical Data Format version 5 (HDF5), the most popular parallel I/O library for scientific applications. HDF5 is one of the most used libraries at the leadership computing facilities for performing parallel I/O on existing HPC systems. The state-of-the-art features we describe include: Virtual Object Layer (VOL), Data Elevator, asynchronous I/O, full-featured single-writer and multiple-reader (Full SWMR), and parallel querying. In this paper, we introduce these features, their implementations, and the performance and feature benefits to applications and other libraries.
The recent progress in parallel hardware architectures with deeper vector pipelines or many-cores technologies brings opportunities for HEP experiments to take advantage of SIMD and SIMT computing models. Launched in 2013, the GeantV project studies performance gains in propagating multiple particles in parallel, improving instruction throughput and data locality in HEP event simulation on modern parallel hardware architecture. Due to the complexity of geometry description and physics algorithms of a typical HEP application, performance analysis is indispensable in identifying factors limiting parallel execution. In this report, we will present design considerations and preliminary computing performance of GeantV physics models on coprocessors (Intel Xeon Phi and NVidia GPUs) as well as on mainstream CPUs.
A memory device receives a plurality of read commands and/or write commands in parallel. The memory device transmits data corresponding to respective read commands on respective portions of a data bus and receives data corresponding to respective write commands on respective portions of the data bus. The memory device includes I/O logic to receive the plurality of read commands in parallel, to transmit the data corresponding to the respective read commands on respective portions of the data bus, and to receive the data corresponding to the respective write commands on respective portions of the data bus.
Formulas for incremental or parallel computation of second order central moments have long been known, and recent extensions of these formulas to univariate and multivariate moments of arbitrary order have been developed. Such formulas are of key importance in scenarios where incremental results are required and in parallel and distributed systems where communication costs are high. We survey these recent results, and improve them with arbitrary-order, numerically stable one-pass formulas which we further extend with weighted and compound variants. We also develop a generalized correction factor for standard two-pass algorithms that enables the maintenance of accuracy over nearly the full representable range of the input, avoiding the need for extended-precision arithmetic. We then empirically examine algorithm correctness for pairwise update formulas up to order four as well as condition number and relative error bounds for eight different central moment formulas, each up to degree six, to address the trade-offs between numerical accuracy and speed of the various algorithms. Finally, we demonstrate the use of the most elaborate among the above mentioned formulas, with the utilization of the compound moments for a practical large-scale scientific application.
Large-scale parallel applications with complex global data dependencies beyond those of reductions pose significant scalability challenges in an asynchronous runtime system. Internodal challenges include identifying the all-to-all communication of data dependencies among the nodes. Intranodal challenges include gathering together these data dependencies into usable data objects while avoiding data duplication. This paper addresses these challenges within the context of a large-scale, industrial coal boiler simulation using the Uintah asynchronous many-task runtime system on GPU architectures. We show significant reduction in time spent analyzing data dependencies through refinements in our dependency search algorithm. Multiple task graphs are used to eliminate subsequent analysis when task graphs change in predictable and repeatable ways. Using a combined data store and task scheduler redesign reduces data dependency duplication ensuring that problems fit within host and GPU memory. Furthermore, these modifications did not require any changes to application code or sweeping changes to the Uintah runtime system. We report results running on the DOE Titan system on 119K CPU cores and 7.5K GPUs simultaneously. Our solutions can be generalized to other task dependency problems with global dependencies among thousands of nodes which must be processed efficiently at large scale.
The nonlocal implementation of the dispersive optical model (DOM) provides all the ingredients for distorted-wave impulse-approximation (DWIA) calculations of the (e,e'p) reaction. It provides both the overlap function, including its normalization, and the outgoing proton distorted wave. This framework is applied to describe the knockout of a proton from the 0d3/2 and 1s1/2 orbitals in 40 Ca with fixed normalizations of 0.71 and 0.60, respectively. Data were obtained in parallel kinematics for three outgoing proton energies: 70, 100, and 135 MeV. Agreement with the data is as good as, or better than, previous descriptions employing local optical potentials and overlap functions from Woods-Saxon potentials—both with standard nonlocality corrections—whose normalization (spectroscopic factor) and radius were fitted to the data. The present analysis suggests that slightly larger spectroscopic factors are obtained when nonlocal optical potentials are employed than those generated with local potentials. The results further suggest that the chosen kinematical window around 100 MeV proton energy provides the best and cleanest method to employ the DWIA for the analysis of this reaction. The conclusion that substantial ground-state correlations cannot be ignored when describing a closed-shell atomic nucleus is therefore confirmed in detail. To reach these conclusions, it is essential to have a complete description of the nucleon single-particle propagator that accounts for all elastic nucleon-scattering observables in a wide energy domain up to 200 MeV. Lastly, the current nonlocal implementation of the DOM fulfills this requirement.
Not Available
In this work, we search the complete Hubble Frontier Field data set of Abell 2744 and its parallel field for $z\sim 10$ sources to further refine the evolution of the cosmic star formation rate density (SFRD) between $z\sim 8$ and $z\sim 10$. We independently confirm two images of the recently discovered triply imaged $z\sim 9.8$ source by Zitrin et al. and set an upper limit for similar $z\sim 10$ galaxies with red colors of ${J}_{125}-{H}_{160}\gt 1.2$ in the parallel field of Abell 2744. We utilize extensive simulations to derive the effective selection volume of Lyman-break galaxies at $z\sim 10$, both in the lensed cluster field and in the adjacent parallel field. Particular care is taken to include position-dependent lensing shear to accurately account for the expected sizes and morphologies of highly magnified sources. We show that both source blending and shear reduce the completeness at a given observed magnitude in the cluster, particularly near the critical curves. These effects have a significant, but largely overlooked, impact on the detectability of high-redshift sources behind clusters, and substantially reduce the expected number of highly magnified sources. The detections and limits from both pointings result in an SFRD which is consistent within the uncertainties with previous estimates at $z\sim 10$ from blank fields. The combination of these new results with all other estimates is also consistent with a rapidly declining SFRD in the 170 Myr from $z\sim 8$ to $z\sim 10$ as predicted by cosmological simulations and dark-matter halo evolution in ΛCDM. Once biases introduced by magnification-dependent completeness are accounted for, the full six cluster and parallel Frontier Field program will be an extremely powerful new data set to probe the evolution of the galaxy population at $z\gt 8$ before the advent of the James Webb Space Telescope.
In this study, we propose a general algorithm for nonconforming adaptive mesh refinement (AMR) of unstructured meshes in high-order finite element codes. Our focus is on $h$-refinement with a fixed polynomial order. The algorithm handles triangular, quadrilateral, hexahedral, and prismatic meshes of arbitrarily high-order curvature, for any order finite element space in the de Rham sequence. We present a flexible data structure for meshes with hanging nodes and a general procedure to construct the conforming interpolation operator, both in serial and in parallel. The algorithm and data structure allow anisotropic refinement of tensor product elements in two and three dimensions and support unlimited refinement ratios of adjacent elements. We report numerical experiments verifying the correctness of the algorithms and perform a parallel scaling study to show that we can adapt meshes containing billions of elements and run efficiently on 393,000 parallel tasks. Finally, we illustrate the integration of dynamic AMR into a high-order Lagrangian hydrodynamics solver.
Purpose: The authors introduce a state-of-the-art all-optical clinical diffuse optical tomography (DOT) imaging instrument which collects spatially dense, multispectral, frequency-domain breast data in the parallel-plate geometry. Methods: The instrument utilizes a CCD-based heterodyne detection scheme that permits massively parallel detection of diffuse photon density wave amplitude and phase for a large number of source–detector pairs (10{sup 6}). The stand-alone clinical DOT instrument thus offers high spatial resolution with reduced crosstalk between absorption and scattering. Other novel features include a fringe profilometry system for breast boundary segmentation, real-time data normalization, and a patient bed design which permits both axial and sagittal breast measurements. Results: The authors validated the instrument using tissue simulating phantoms with two different chromophore-containing targets and one scattering target. The authors also demonstrated the instrument in a case study breast cancer patient; the reconstructed 3D image of endogenous chromophores and scattering gave tumor localization in agreement with MRI. Conclusions: Imaging with a novel parallel-plate DOT breast imager that employs highly parallel, high-resolution CCD detection in the frequency-domain was demonstrated.
The American Recovery and Reinvestment Act stimulus funding of 2009 for smart grid projects resulted in the tripling of smart meters deployment. In 2012, the Green Button initiative provided utility customers with access to their real-time1 energy usage. The availability of finely granular data provides an enormous potential for energy data analytics and energy benchmarking. The sheer volume of time-series utility data from a large number of buildings also poses challenges in data collection, quality control, and database management for rigorous and meaningful analyses. In this paper, we will describe a building portfolio-level data analytics tool for operational optimization, business investment and policy assessment using 15-minute to monthly intervals utility data. The analytics tool is developed on top of the U.S. Department of Energy’s Standard Energy Efficiency Data (SEED) platform, an open source software application that manages energy performance data of large groups of buildings. To support the significantly large volume of granular interval data, we integrated a parallel time-series database to the existing relational database. The time-series database improves on the current utility data input, focusing on real-time data collection, storage, analytics and data quality control. The fully integrated data platform supports APIs for utility apps development by third party software developers. These apps will provide actionable intelligence for building owners and facilities managers. Unlike a commercial system, this platform is an open source platform funded by the U.S. Government, accessible to the public, researchers and other developers, to support initiatives in reducing building energy consumption.