Search NASA⌕ Search

SEARCH · Search NASA

Results for “parallel systems”

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 1,027 records · Page 57

Divide and Conquer (DC) BLAST: fast and easy BLAST execution within HPC environments

Bioinformatics is currently faced with very large-scale data sets that lead to computational jobs, especially sequence similarity searches, that can take absurdly long times to run. For example, the National Center for Biotechnology Information (NCBI) Basic Local Alignment Search Tool (BLAST and BLAST+) suite, which is by far the most widely used tool for rapid similarity searching among nucleic acid or amino acid sequences, is highly central processing unit (CPU) intensive. While the BLAST suite of programs perform searches very rapidly, they have the potential to be accelerated. In recent years, distributed computing environments have become more widely accessible and used due to the increasing availability of high-performance computing (HPC) systems. Therefore, simple solutions for data parallelization are needed to expedite BLAST and other sequence analysis tools. However, existing software for parallel sequence similarity searches often requires extensive computational experience and skill on the part of the user. In order to accelerate BLAST and other sequence analysis tools, Divide and Conquer BLAST (DCBLAST) was developed to perform NCBI BLAST searches within a cluster, grid, or HPC environment by using a query sequence distribution approach. Scaling from one (1) to 256 CPU cores resulted in significant improvements in processing speed. Thus, DCBLAST dramatically accelerates the execution of BLAST searches using a simple, accessible, robust, and parallel approach. DCBLAST works across multiple nodes automatically and it overcomes the speed limitation of single-node BLAST programs. DCBLAST can be used on any HPC system, can take advantage of hundreds of nodes, and has no output limitations. Thus, this freely available tool simplifies distributed computation pipelines to facilitate the rapid discovery of sequence similarities between very large data sets.

59 BASIC BIOLOGICAL SCIENCES↗

Structural Simluation Toolkit (SST) v.12.0

The Structural Simulation Toolkit (SST) was developed to explore innovations in highly concurrent computing systems where the instruction set architecture (ISA), micro-architecture, and memory interact with the programming model and communications system. The package provides a fully modular design for extensive exploration of an individual system parameter as well as a parallel simulation environment based on message passing interface (MPI) which enable a high level of performance as well as the ability to look at large systems. 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.

Rodrigues, ArunF.↗

Ni and CoO spin cantings induced by Fe layer in Ni/CoO/Fe/vicinal MgO(001)

Using element-resolved x-ray magnetic circular dichroism and x-ray magnetic linear dichroism measurements, we studied Ni/CoO/vicinal MgO(001) and Ni/CoO/Fe/vicinal MgO(001) systems at 350 and 78 K. Above the CoO Néel temperature, the Ni magnetization is fully in plane and parallel to the atomic steps in both systems due to step-induced magnetic anisotropy. Below the CoO Néel temperature, the CoO spins in Ni/CoO/vicinal MgO(001) are fully in plane and parallel to the atomic steps and the Ni magnetization is fully in plane and perpendicular to the atomic steps due to the 90° Ni/CoO magnetic coupling. The CoO spins in Ni/CoO/Fe/vicinal MgO(001), however, develop an out-of-plane canted spin component in addition to the in-plane component parallel to the atomic steps. Consequently, the Ni magnetization is canted towards the out-of-plane direction by an appreciable angle. Finally, photoemission electron microscopy imaging shows a 90° interfacial magnetic coupling at both the Ni/CoO and the CoO/Fe interfaces and an absence of a direct interlayer coupling, showing that the Ni spin canting is due to its coupling to the canted CoO spin components which is caused by the underlying ferromagnetic Fe layer in Ni/CoO/Fe/vicinal MgO(001).

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

Closed-Loop Error Learning Control for Uncertain Nonlinear Systems With Experimental Validation on a Mobile Robot

This article develops a closed-loop error learning control (CLELC) algorithm for feedback linearizable systems with experimental validation on a mobile robot. Traditional feedback and feedforward controllers are designed based on the nominal model by using the feedback linearization control (FLC) method. Then, an intelligent controller is designed based on the sliding mode learning algorithm that utilizes closed-loop error dynamics to learn the system behavior. The controllers are working in parallel, and the intelligent controller can gradually replace the feedback controller from the control of the system. In addition to the stability of the sliding mode learning algorithm, the closed-loop stability of an nth-order feedback linearizable system is proven. The simulation results demonstrate that the CLELC algorithm can improve control performance (e.g., smaller rise time, settling time, and overshoot) in the absence of uncertainties, and also provides robust control performance in the presence of uncertainties as compared to traditional FLC method. To test the efficiency and efficacy of the CLELC algorithm, the trajectory tracking problem of a tracked mobile robot is studied in real time. The experimental results demonstrate that the CLELC algorithm ensures highly accurate trajectory tracking performance than traditional FLC method.

97 MATHEMATICS AND COMPUTING↗

Characterizing and Mitigating Work Time Inflation in Task Parallel Programs

Task parallelism raises the level of abstraction in shared memory parallel programming to simplify the development of complex applications. However, task parallel applications can exhibit poor performance due to thread idleness, scheduling overheads, and work time inflation – additional time spent by threads in a multithreaded computation beyond the time required to perform the same work in a sequential computation. We identify the contributions of each factor to lost efficiency in various task parallel OpenMP applications and diagnose the causes of work time inflation in those applications. Increased data access latency can cause significant work time inflation in NUMA systems. Our locality framework for task parallel OpenMP programs mitigates this cause of work time inflation. Our extensions to the Qthreads library demonstrate that locality-aware scheduling can improve performance up to 3X compared to the Intel OpenMP task scheduler.

Olivier, Stephen L.↗

Comparison of entropy production rates in two different types of self-organized flows: Benard convection and zonal flow

Two different types of self-organizing and sustaining ordered motion in fluids or plasmas--one is a Benard convection (or streamer) and the other is a zonal flow--have been compared by introducing a thermodynamic phenomenological model and evaluating the corresponding entropy production rates (EP). These two systems have different topologies in their equivalent circuits: the Benard convection is modeled by parallel connection of linear and nonlinear conductances, while the zonal flow is modeled by series connection. The ''power supply'' that drives the systems is also a determinant of operating modes. When the energy flux is a control parameter (as in usual plasma experiments), the driver is modeled by a constant-current power supply, and when the temperature difference between two separate boundaries is controlled (as in usual computational studies), the driver is modeled by a constant-voltage power supply. The parallel (series)-connection system tends to minimize (maximize) the total EP when a constant-current power supply drives the system. This minimum/maximum relation flips when a constant-voltage power supply is connected.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

DeepHyper: Asynchronous Hyperparameter Search for Deep Neural Networks

Hyperparameters employed by deep learning (DL) methods play a substantial role in the performance and reliability of these methods in practice. Unfortunately, finding performance-optimizing hyperparameter settings is a notoriously difficult task. Hyperparameter search methods typically have limited production-strength implementations or do not target scalability within a highly parallel machine, portability across different machines, experimental comparison between different methods, and tighter integration with workflow systems. In this paper, we present DeepHyper, a Python package that provides a common interface for the implementation and study of scalable hyperparameter search methods. It adopts the Balsam workflow system to hide the complexities of running large numbers of hyperparameter configurations in parallel on high-performance computing (HPC) systems. We implement and study asynchronous model-based search methods that consist of sampling a small number of input hyperparameter configurations and progressively fitting surrogate models over the input-output space until exhausting a user-defined budget of evaluations. We evaluate the efficacy of these methods relative to approaches such as random search, genetic algorithms, Bayesian optimization, and hyperband on DL benchmarks on CPU- and GPU-based HPC systems.

Balaprakash, Prasanna↗

Statistical Fault Detection for Parallel Applications with AutomaDeD

Today's largest systems have over 100,000 cores, with million-core systems expected over the next few years. The large component count means that these systems fail frequently and often in very complex ways, making them difficult to use and maintain. While prior work on fault detection and diagnosis has focused on faults that significantly reduce system functionality, the wide variety of failure modes in modern systems makes them likely to fail in complex ways that impair system performance but are difficult to detect and diagnose. This paper presents AutomaDeD, a statistical tool that models the timing behavior of each application task and tracks its behavior to identify any abnormalities. If any are observed, AutomaDeD can immediately detect them and report to the system administrator the task where the problem began. This identification of the fault's initial manifestation can provide administrators with valuable insight into the fault's root causes, making it significantly easier and cheaper for them to understand and repair it. Our experimental evaluation shows that AutomaDeD detects a wide range of faults immediately after they occur 80% of the time, with a low false-positive rate. Further, it identifies weaknesses of the current approach that motivate future research.

99 GENERAL AND MISCELLANEOUS↗

AnalyzeThat: A Programmable Shared-Memory System for an Array of Processing-In-Memory Devices

Processing In Memory (PIM), the concept of integrating processing directly with memory, has been attracting a lot of attention since PIM can assist in overcoming the throughput limitation caused by data movement between CPU and memory. The challenge, however, is that it requires the programmers to have a deep understanding of the PIM architecture to maximize the benefits such as data locality and parallel thread execution on multiple PIM devices. In this study, we present AnalyzeThat, a programmable shared-memory system for parallel data processing with PIM devices. Thematic to AnalyzeThat is a rich PIM-Aware Data Structure (PADS), which is an encapsulation that integrally ties together the data, the analysis tasks and the runtime needed to interface with the PIM device array. The PADS abstraction provides (i) a key-value data container that allows programmers to easily store data on multiple PIMs, (ii) a suite of parallel operations with which users can easily implement data analysis applications, and (iii) a runtime, hidden to programmers, which provides the mechanisms needed to overlay both the data and the tasks on the PIM device array in an intelligent fashion, based on PIM-specific information collected from the hardware. We have developed a PIM emulation framework called AnalyzeThat. Our experimental evaluation with representative data analytics applications suggests that the proposed system can significantly reduce the PIM programming effort without losing its technology benefits.

97 MATHEMATICS AND COMPUTING↗

A programmable shared-memory system for an array of processing-in-memory devices

Processing in memory (PIM), the concept of integrating processing directly with memory has been attracting a lot of attention, since PIM can assist in overcoming the throughput limitation caused by data movement between CPU and memory. The challenge, however, is that it requires the programmers to have a deep understanding of the PIM architecture to maximize the benefits such as data locality and parallel thread execution on multiple PIM devices. In this study, we present AnalyzeThat, a programmable shared-memory system for parallel data processing with PIM devices. Thematic to AnalyzeThat is a rich PIM-aware data structure (PADS), which is an encapsulation that integrally ties together the data, the analysis tasks and the runtime needed to interface with the PIM device array. The PADS abstraction provides (i) a sophisticated key-value data container that allows programmers to easily store data on multiple PIMs, (ii) a suite of parallel operations with which users can easily implement data analysis applications, and (iii) a runtime, hidden to programmers, which provides the mechanisms needed to overlay both the data and the tasks on the PIM device array in an intelligent fashion, based on PIM-specific information collected from the hardware. We have developed a PIM emulation framework called AnalyzeThat. In conclusion, our experimental evaluation with representative data analytics applications suggests that the proposed system can significantly reduce the PIM programming effort without losing its technology benefits.

97 MATHEMATICS AND COMPUTING↗

Taming parallel I/O complexity with auto-tuning

We present an auto-tuning system for optimizing I/O performance of HDF5 applications and demonstrate its value across platforms, applications, and at scale. The system uses a genetic algorithm to search a large space of tunable parameters and to identify effective settings at all layers of the parallel I/O stack. The parameter settings are applied transparently by the auto-tuning system via dynamically intercepted HDF5 calls. To validate our auto-tuning system, we applied it to three I/O benchmarks (VPIC, VORPAL, and GCRM) that replicate the I/O activity of their respective applications. We tested the system with different weak-scaling configurations (128, 2048, and 4096 CPU cores) that generate 30 GB to 1 TB of data, and executed these configurations on diverse HPC platforms (Cray XE6, IBM BG/P, and Dell Cluster). In all cases, the auto-tuning framework identified tunable parameters that substantially improved write performance over default system settings. In conclusion, we consistently demonstrate I/O write speedups between 2x and 100x for test configurations.

97 MATHEMATICS AND COMPUTING↗

Effect of longitudinal magnetic fields on a simulated in-line 6 MV linac

Purpose: Linac-magnetic resonance (MR) systems have been proposed in order to achieve real-time image guided radiotherapy. The design of a new linac-MR system with the in-line 6 MV linac generating x-rays along the symmetry axis of an open MR imager is outlined. This new design allows for a greater MR field strength to achieve better quality images while reducing hot and cold spots in treatment planning. An investigation of linac's performance in the longitudinal fringe magnetic fields of the MR imager is given. Methods: The open MR imager fringe magnetic field was modeled using the analytic solution of the magnetic field generated from current carrying loops. The derived solution was matched to the magnetic fringe field isolines provided for a 0.5 T open MR imager through Monte Carlo optimization. The optimized field solution was then added to the previously validated 6 MV linac simulation to quantify linac's performance in the fringe magnetic field of a 0.5 T MR imager. To further the investigation, linac's performance in large fringe fields expected from other imagers was investigated through the addition of homogeneous longitudinal fields. Results: The Monte Carlo optimization of the analytic current loop solution provided good agreement with the magnetic fringe field isolines supplied by the manufacturer. The range of magnetic fields the linac is expected to experience when coupled to the 0.5 T MR imager was determined to be from 0.0022 to 0.011 T (as calculated at the electron gun cathode). The effect of the longitudinal magnetic field on the electron beam was observed to be only in the electron gun. The longitudinal field changed the electron gun optics, affecting beam characteristics, such as a slight increase in the injection current and beam diameter, and an increasingly nonlaminar transverse phase space. Although the target phase space showed little change in its energy spectrum from the altered injection phase space, a reduction in the target current and spatial distribution peak intensity was observed. Despite these changes, the target phase space had little effect on the depth dose curves or dose profiles calculated for a 40x40 cm{sup 2} field at 1.5 cm depth. At longitudinal fields larger than 0.012 T, a drastic reduction in the injection current from the electron gun was observed due to a large fraction of electrons striking the anode. This further reduced the target current, which reached a minimum of 28{+-}2 mA at 0.06 T. A slow increase in the injection and target currents was observed at fields larger than 0.06 T due to greater beam collimation in the anode beam tube. Conclusions: In an effort to achieve higher quality images and a reduction in hot and cold spots in the treatment plan, a parallel configuration linac-MR system is presented. The longitudinal magnetic fields of the MR imager caused large beam losses within the electron gun. These losses may be eliminated through a redesign of the electron gun optics incorporating a longitudinal magnetic field, or through magnetic shielding, which has already been proven successful for the transverse configuration.

61 RADIATION PROTECTION AND DOSIMETRY↗

Water–Hydrocarbon Interactions in Anionic Pyrene Monohydrate

Interactions between water and polycyclic aromatic hydrocarbons are essential in many aspects of chemistry, from interstellar and atmospheric processes to interfacial hydrophobicity and wetting phenomena. Despite their growing importance, the intermolecular potentials of the water-hydrocarbon interactions are underdeveloped compared to water-water potentials, and there are similarly few experimental probes that are sensitive to the details of the water-hydrocarbon potential. We present a combined experimental and computational study of anionic pyrene monohydrate, one of the simplest water/hydrocarbon clusters. The action spectrum in the OH region of the mass-selected cluster ion provides a rigorous benchmark for intermolecular potentials and computational methodologies. We identify missing intermolecular interactions and shortcomings in conventional dynamics calculations by comparing experimental data to density functional theory and classical molecular dynamics calculations. Kinetic trapping is prevalent, even for one water molecule and one pyrene molecule, leading to slow equilibration in conventional molecular dynamics calculations, even on nanosecond timescales and at low temperatures (50 K). At constant energy, temperature fluctuations for the pair of molecules are substantial. Immersing the system in a bath of soft spheres and employing parallel tempering alleviates kinetic trapping and dampens temperature fluctuations, bringing the system closer to the thermodynamic limit. With such augmented sampling, a simple, flexible water model reproduces the linewidth and the asymmetric broadening of the symmetric OH stretching mode, which we assign to spectral diffusion. In the OH stretching region, dynamics calculations predict a more intense antisymmetric peak than experiments observe but do not predict the bimodal split symmetric peak that the experiments show. Furthermore, our work suggests that electronic polarization, missing in the empirical force field, is responsible for the first discrepancy and that quantum nuclear effects, captured neither in density functional theory nor in classical dynamics, may be responsible for the second.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

PDSLin User Guide

This document describes a software package which implements a parallel hybrid (direct/iterative) linear solver based on the Schur complement method for solving a general sparse linear system of equations. The package is named parallel domain decomposition Schur complement based linear solver, or PDSLin in short. We give a brief description of the algorithm, installation, calling sequences, and data structures of PDSLin.

97 MATHEMATICS AND COMPUTING↗

Active Learning for Metamaterial Optimization on HPC and QC Integrated Systems

Active learning algorithms, integrating machine learning, quantum computing and optics simulation in an iterative loop, offer a promising approach to optimizing metamaterials. However, these algorithms can face difficulties in optimizing highly complex structures due to computational limitations. High-performance computing (HPC) and quantum computing (QC) integrated systems can address these issues by enabling parallel computing. In this study, we develop an active learning algorithm working on HPC-QC integrated systems. We evaluate the performance of optimization processes within active learning (i.e., training a machine learning model, problem-solving with quantum computing, and evaluating optical properties through wave-optics simulation) for highly complex metamaterial cases. Our results showcase that utilizing multiple cores on the integrated system can significantly reduce computational time, thereby enhancing the efficiency of optimization processes. Therefore, we expect that leveraging HPC-QC integrated systems helps effectively tackle large-scale optimization challenges in general.

Kim, Seongmin↗

Ethanol Dehydration to Ethylene in a Stratified Autothermal Millisecond Reactor

The concurrent decomposition and deoxygenation of ethanol was accomplished in a stratified reactor with 50-80 ms contact times. The stratified reactor comprised an upstream oxidation zone that contained Pt-coated Al(2)O(3) beads and a downstream dehydration zone consisting of H-ZSM-5 zeolite films deposited on Al(2)O(3) monoliths. Ethanol conversion, product selectivity, and reactor temperature profiles were measured for a range of fuel:oxygen ratios for two autothermal reactor configurations using two different sacrificial fuel mixtures: a parallel hydrogen-ethanol feed system and a series methane-ethanol feed system. Increasing the amount of oxygen relative to the fuel resulted in a monotonic increase in ethanol conversion in both reaction zones. The majority of the converted carbon was in the form of ethylene, where the ethanol carbon-carbon bonds stayed intact while the oxygen was removed. Over 90% yield of ethylene was achieved by using methane as a sacrificial fuel. These results demonstrate that noble metals can be successfully paired with zeolites to create a stratified autothermal reactor capable of removing oxygen from biomass model compounds in a compact, continuous flow system that can be configured to have multiple feed inputs, depending on process restrictions.

Skinner, MJ↗

25‐4: Late‐News Paper: Xerographic MicroAssembly Printer for LEDs and Beyond

We are building a deterministic massively parallel sorting, assembly and transfer system for microLEDs and other applications. It uses programmable directed electrostatic assembly and transfer (stamp or roller). Small chips (10 um – 500 um), fine registration (<1um, <1°), heterogeneity, and 162 chiplet prints over an inch are demonstrated.

Chow, Eugene M.↗

Effect of Neoadjuvant and Adjuvant Therapy of Experimental Breast Cancer on the Structure of Mesenteric Lymph Nodes

The morphometric analysis of mesenteric lymph nodes was carried out in female Wistar rats with chemically induced breast cancer. In control rats with untreated breast cancer, the volume of the system of sinuses increased in parallel with the appearance of morphological signs of suppression of cell-mediated immunity, inhibition of humoral immunity, and macrophage reaction. Against the background of chemotherapy, we observed a decrease in the volume of paracortex and lymphoid nodules, suppression of proliferative activity of lymphoid cells in paracortical and B-cell zone, and a decrease in macrophage content. After resection of breast cancer followed by chemotherapy course, lymph transport activation, widening of the paracortex, enhanced proliferative activity of cells in the paracortex and B-cell zone, and reduced volumes of lymphoid nodules with and without germinal centers and medullary substance were revealed in comparison with rats subjected neoadjuvant chemotherapy.

60 APPLIED LIFE SCIENCES↗