Search NASA⌕ Search

SEARCH · Search NASA

Results for “parallel, scaling”

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 1,081 records · Page 60
All sourcesNASA NTRSDOE OSTI
NASA NTRS1987

Multichannel optical correlator/convolver utilizing the magnetooptic spatial light modulator

The paper describes a multichannel correlator/convolver architecture utilizing an acoustooptic light modulator for the one dimensional channel and a magnetooptic spatial light modulator (MOSLM) for the second two-dimensional parallel channel. The MOSLM allows greater parallelism to be implemented in this correlator/convolver design than was previously reported. The implementation of 24 parallel channels with a 48 x 48 device is demonstrated. Experimental data are presented and the ways of increasing the number of parallel channels using commercially available MOSLMs and other previously discussed techniques, such as frequency multiplexing, are discussed. It is shown that over 2000 parallel channels are possible at 32-bit accuracy. A technique for obtaining a limited gray scale is also discussed.

Davis, Jeffrey A.↗
NASA NTRS1992

Polarization of the far-infrared emission from the thermal filaments of the Galactic center arc

The polarization of the 100 micron continuum emission has been measured at 14 positions in the dense, warm molecular cloud associated with the arched filaments, or the 'bridge', of the radio arc near the Galactic center. At all positions the percent polarization is found to be quite large, ranging up to 6.5 percent. The polarization is interpreted in terms of thermal emission by magnetically aligned dust grains. The directions of the polarization vectors then indicate that the magnetic field is (1) parallel to the long dimension of the thermal radio filaments, and (2) very uniform on scales of 1-10 pc. Of several explanations for the inferred field geometry, the simplest is that it results from the unusually large dynamical shear in the emitting cloud.

Morris, M.↗
NASA NTRS2002

Load Balancing Strategies for Multi-Block Overset Grid Applications

The multi-block overset grid method is a powerful technique for high-fidelity computational fluid dynamics (CFD) simulations about complex aerospace configurations. The solution process uses a grid system that discretizes the problem domain by using separately generated but overlapping structured grids that periodically update and exchange boundary information through interpolation. For efficient high performance computations of large-scale realistic applications using this methodology, the individual grids must be properly partitioned among the parallel processors. Overall performance, therefore, largely depends on the quality of load balancing. In this paper, we present three different load balancing strategies far overset grids and analyze their effects on the parallel efficiency of a Navier-Stokes CFD application running on an SGI Origin2000 machine.

Djomehri, M. Jahed↗
NASA NTRS2004

Measurements of Infrared and Acoustic Source Distributions in Jet Plumes

The aim of this investigation was to use the linear phased array (LPA) microphones and infrared (IR) imaging to study the effects of advanced nozzle-mixing techniques on jet noise reduction. Several full-scale engine nozzles were tested at varying power cycles with the linear phased array setup parallel to the jet axis. The array consisted of 16 sparsely distributed microphones. The phased array microphone measurements were taken at a distance of 51.0 ft (15.5 m) from the jet axis, and the results were used to obtain relative overall sound pressure levels from one nozzle design to the other. The IR imaging system was used to acquire real-time dynamic thermal patterns of the exhaust jet from the nozzles tested. The IR camera measured the IR radiation from the nozzle exit to a distance of six fan diameters (X/D(sub FAN) = 6), along the jet plume axis. The images confirmed the expected jet plume mixing intensity, and the phased array results showed the differences in sound pressure level with respect to nozzle configurations. The results show the effects of changes in configurations to the exit nozzles on both the flows mixing patterns and radiant energy dissipation patterns. By comparing the results from these two measurements, a relationship between noise reduction and core/bypass flow mixing is demonstrated.

Agboola, Femi A.↗
NASA NTRS2003

Parallel Subconvolution Filtering Architectures

These architectures are based on methods of vector processing and the discrete-Fourier-transform/inverse-discrete- Fourier-transform (DFT-IDFT) overlap-and-save method, combined with time-block separation of digital filters into frequency-domain subfilters implemented by use of sub-convolutions. The parallel-processing method implemented in these architectures enables the use of relatively small DFT-IDFT pairs, while filter tap lengths are theoretically unlimited. The size of a DFT-IDFT pair is determined by the desired reduction in processing rate, rather than on the order of the filter that one seeks to implement. The emphasis in this report is on those aspects of the underlying theory and design rules that promote computational efficiency, parallel processing at reduced data rates, and simplification of the designs of very-large-scale integrated (VLSI) circuits needed to implement high-order filters and correlators.

Gray, Andrew A.↗
NASA NTRS2020

X-HAB 2020: AR Field Treks Summary and Conclusions

As part of the FY20 X-Hab Challenge, BLiSS sought to create an Augmented Reality (AR) toolkit to help with analog field trek operations under the supervision of the Solar System Exploration Research Virtual Institute (SSERVI). These treks are operational and technical demonstrations at space-like destinations on Earth to test current extra-vehicular activity (EVA) techniques. While BLiSS as an organization has experience studying operational tasks such as this, it has never developed AR software at this scale. For that reason, another team at the University was brought on to work in parallel. The Collaborative Lab for Advancing Work in Space (CLAWS) is a veteran group of the NASA Spacesuit User Interface Technologies for Students (SUITS) challenge in which Hololens displays for astronauts are created within a year. The operational and technological pairing was ideally suited for tackling this problem. The team divided its responsibilities so that BLiSS would handle the research required to shape the project. As this deliverable had an end user, it was decided that interviewing these field geologists and operations specialists would provide the best insight. These interviews paired with literature review would reveal niche applications for AR that remained within feasible bounds. These science-driven EVAs in unknown terrain require more flexible tools than the current generation of EVA assistants. Rather than focus on sequential instructions, there instead needs to be a broad toolkit that's only called upon in specific instances. This AR Toolkit for Lunar Astronauts and Scientists (ATLAS) became the development goal of the project: create a non-intrusive assembly of tools that could be accessed in AR on the field. The current ATLAS design makes use of a geospatially and temporally annotated eld note system called GeoNotes. This allows for data to be collected and coordinated in a way that's synchronized across time, space, and different users. A Mission Control Center (MCC) and Mobile Support Equipment (MSE) were all needed to transport the AR headset into the field with the user. A network infrastructure was designed and set up within the University to enable this functionality. The software is based on a Protocol-Module structure that allows for modular development of each capability. A Protocol Manager coordinates different protocols that make use of modules. Each module tackles a different individual task while the protocol puts each one to use. The protocol manager coordinates when these are called to be used. This software is hosted on a head-mounted display (HMD) with the MCC acting as support from afar. While the software would be unit-tested at each level and each hardware component verified, a final demonstration would serve to prove the system's capabilities: an analog field trek. The team would prepare to support a user in a remote location from the MCC back at the University. A local area near campus would be tested before going out to do sample field geology further away. This unfortunately became impossible with the arrival of COVID-19. Access to all of the facilities to complete the project as planned were shut down. Our team was scattered across the globe and forced to complete the rest virtually. Adjustments were made to produce a small virtual concept in Adobe XD in the meantime. Even digital surveys were created based on the NASA task-load index (TLX) originally intended for testing actual users. The goal shifted towards completing software and getting feedback on the user interfaces (UI) and user experiences (UX). This team has reformed in response to COVID and its focus has shifted to what can be done remotely. There is still an intention to finish the original deliverable described in this report. The work has been expanded beyond the original X-Hab challenge and has instead become its own research e ort to be continued afterwards. This report collects the processes and knowledge gained from a year of studying and working at this problem with two teams. It should preserve it for the time until the world returns to normal and work can resume. CLAWS will be taking over full responsibility from that point forward, eventually surpassing the original needs of the project. While this document captures the work done towards an eventual end, the CLAWS team has written their own proposal alongside it. It outlines a new future for ATLAS beyond X-Hab, BLiSS, and hopefully beyond COVID-19. This project began as a vague goal hoping to place a new technology into the unique setting of exploration science. The project has since comfortably taken root and will hopefully bloom over the next year.

Alex Sena↗
DOE OSTI2023

A PIC Bootstrapping Strategy for Exascale CFD-DEM Simulation

The exascale computing era comes with the release of MFIX-Exa, a new code for studying gas-particle fluidization using CFD-DEM and PIC models on massively parallel, heterogeneous high-performance computing architectures. However, as compute resources continue to grow in scale and complexity, so too does the impetus to use them efficiently. Here, we propose a novel bootstrapping method to minimize neglected simulation time and cast the approach more broadly among a growing field of multi-fidelity methods.

Porcu, Roberto↗
DOE OSTI2017

Task-based Parallel Computation of the Density Matrix in Quantum-based Molecular Dynamics using Graph Partitioning

Quantum-based molecular dynamics (QMD) is a highly accurate and transferable method for material science simulations. However, the time scales and system sizes accessible to QMD are typically limited to picoseconds and a few hundred atoms. These constraints arise due to expensive self-consistent ground-state electronic structure calculations that can often scale cubically with the number of atoms. Linearly scaling methods depend on computing the density matrix P from the Hamiltonian matrix H by exploiting the sparsity in both matrices. The second-order spectral projection (SP2) algorithm is an O(N) algorithm that computes P with a sequence of 40 - 50 matrix-matrix multiplications. In this paper, we present task-based implementations of a recently developed data-parallel graph-based approach to the SP2 algorithm, G-SP2. We represent the density matrix P as an undirected graph and use graph partitioning techniques to divide the computation into smaller independent tasks. The partitions thus obtained are generally not of equal size and give rise to undesirable load imbalances in standard MPI-based implementations. This load-balancing challenge can be mitigated by dynamically scheduling parallel computations at runtime using task-based programming models. We develop task-based implementations of the data-parallel G-SP2 algorithm using both Intel's Concurrent Collections (CnC) as well as the Charm++ programming model and evaluate these implementations for future use. Scaling and performance results of our implementations are investigated for representative segments of QMD simulations for solvated protein systems containing more than 10,000 atoms.

97 MATHEMATICS AND COMPUTING↗
NASA NTRS2003

Hybrid VLSI/QCA Architecture for Computing FFTs

A data-processor architecture that would incorporate elements of both conventional very-large-scale integrated (VLSI) circuitry and quantum-dot cellular automata (QCA) has been proposed to enable the highly parallel and systolic computation of fast Fourier transforms (FFTs). The proposed circuit would complement the QCA-based circuits described in several prior NASA Tech Briefs articles, namely Implementing Permutation Matrices by Use of Quantum Dots (NPO-20801), Vol. 25, No. 10 (October 2001), page 42; Compact Interconnection Networks Based on Quantum Dots (NPO-20855) Vol. 27, No. 1 (January 2003), page 32; and Bit-Serial Adder Based on Quantum Dots (NPO-20869), Vol. 27, No. 1 (January 2003), page 35. The cited prior articles described the limitations of very-large-scale integrated (VLSI) circuitry and the major potential advantage afforded by QCA. To recapitulate: In a VLSI circuit, signal paths that are required not to interact with each other must not cross in the same plane. In contrast, for reasons too complex to describe in the limited space available for this article, suitably designed and operated QCAbased signal paths that are required not to interact with each other can nevertheless be allowed to cross each other in the same plane without adverse effect. In principle, this characteristic could be exploited to design compact, coplanar, simple (relative to VLSI) QCA-based networks to implement complex, advanced interconnection schemes.

Fijany, Amir↗
NASA NTRS1991

Automated target recognition and tracking using an optical pattern recognition neural network

The on-going development of an automatic target recognition and tracking system at the Jet Propulsion Laboratory is presented. This system is an optical pattern recognition neural network (OPRNN) that is an integration of an innovative optical parallel processor and a feature extraction based neural net training algorithm. The parallel optical processor provides high speed and vast parallelism as well as full shift invariance. The neural network algorithm enables simultaneous discrimination of multiple noisy targets in spite of their scales, rotations, perspectives, and various deformations. This fully developed OPRNN system can be effectively utilized for the automated spacecraft recognition and tracking that will lead to success in the Automated Rendezvous and Capture (AR&C) of the unmanned Cargo Transfer Vehicle (CTV). One of the most powerful optical parallel processors for automatic target recognition is the multichannel correlator. With the inherent advantages of parallel processing capability and shift invariance, multiple objects can be simultaneously recognized and tracked using this multichannel correlator. This target tracking capability can be greatly enhanced by utilizing a powerful feature extraction based neural network training algorithm such as the neocognitron. The OPRNN, currently under investigation at JPL, is constructed with an optical multichannel correlator where holographic filters have been prepared using the neocognitron training algorithm. The computation speed of the neocognitron-type OPRNN is up to 10(exp 14) analog connections/sec that enabling the OPRNN to outperform its state-of-the-art electronics counterpart by at least two orders of magnitude.

Chao, Tien-Hsin↗
DOE OSTI2022

Towards Low-Overhead Resilience for Data Parallel Deep Learning

Data parallel techniques have been widely adopted both in academia and industry as a tool to enable scalable training of deep learning models. At scale, DL training jobs can fail due to software or hardware bugs, may need to be preempted or terminated due to unexpected events, or may perform suboptimally because they were misconfigured. Under such circumstances, there is a need to recover and/or reconfigure data-parallel DL training jobs on-the-fly, while minimizing the impact on the accuracy of the DNN model and the runtime overhead. In this regard, state-of-art techniques adopted by the HPC community mostly rely on checkpoint-restart, which inevitably leads to loss of progress, thus increasing the runtime overhead. In this paper we explore alternative techniques that exploit the properties of modern deep learning frameworks (overlapping of gradient averaging and weight updates with local gradient computations through pipeline parallelism) to reduce the overhead of resilience/elasticity. To this end we introduce a failure simulation framework and two resilience strategies (immediate mini-batch rollback and lossy forward recovery), which we study compared with checkpoint-restart approaches in a variety of settings in order to understand the trade-offs between the accuracy loss of the DNN model and the runtime overhead.

data-parallel training↗
NASA NTRS2010

Establishing a Robotic, LEO-to-GEO Satellite Servicing Infrastructure as an Economic Foundation for Exploration

The strategy for accomplishing civilian exploration goals and objectives is in the process of a fundamental shift towards a potential new approach called Flexible Path. This paper suggests that a government-industry or public-private partnership in the commercial development of low Earth orbit to geostationary orbit (LEO-to-GEO (LTG)) space, following or in parallel with the commercialization of Earth-to-LEO and International Space Station (ISS) operations, could serve as a necessary, logical step that can be incorporated into the flexible path approach. A LTG satellite-servicing infrastructure and architecture concept is discussed within this new strategic context. The concept consists of a space harbor that serves as a transport facility for a fleet of specialized, fully- or semi-autonomous robotic servicing spacecraft. The baseline, conceptual system architecture is composed of a space harbor equipped with specialized servicer spacecraft; a satellite command, communication, and control system; a parts station; a fuel station or depot; and a fuel/parts replenishment transport. The commercial servicer fleet would consist of several types of spacecraft, each designed with specialized robotic manipulation subsystems to provide services such as refueling, upgrade, repair, inspection, relocation, and removal. The space harbor is conceptualized as an ISS-type, octagonal truss structure equipped with radiation tolerant subsystems. This space harbor would be primarily capable of serving as an operational platform for various commercially owned and operated servicer spacecraft positioned and docked symmetrically on four of the eight sides. Several aspects of this concept are discussed, such as: system-level feasibility in terms of ISS-truss-type infrastructure and subsystems emplacement and maintenance between LEO and GEO; infrastructure components assembly in LEO, derived from ISS assembly experience, and transfer to various higher orbital locations; the evolving Earth-to-orbit (ETO) capability to deliver humans and cargo to LEO for assembly purposes; system architectural definition, optimal orbital parameters, mass estimations, delta velocity ( V) estimations, power and propulsion options, and assessments of various critical technologies. Large-scale, robotic, LTG satellite servicing is considered as an essential economic pre-condition and next parallel or sequential step on the road toward exploration beyond LEO. Such a step might produce the necessary pre-requisite economic value that can be used by future decision makers to justify further investment in exploration beyond LEO.

Horsham, Gary A. P.↗
DOE OSTI2017

Cross-scale efficient tensor contractions for coupled cluster computations through multiple programming model backends

Coupled-cluster methods provide highly accurate models of molecular structure through explicit numerical calculation of tensors representing the correlation between electrons. These calculations are dominated by a sequence of tensor contractions, motivating the development of numerical libraries for such operations. While based on matrix–matrix multiplication, these libraries are specialized to exploit symmetries in the molecular structure and in electronic interactions, and thus reduce the size of the tensor representation and the complexity of contractions. The resulting algorithms are irregular and their parallelization has been previously achieved via the use of dynamic scheduling or specialized data decompositions. We introduce our efforts to extend the Libtensor framework to work in the distributed memory environment in a scalable and energy-efficient manner. We achieve up to 240× speedup compared with the optimized shared memory implementation of Libtensor. We attain scalability to hundreds of thousands of compute cores on three distributed-memory architectures (Cray XC30 and XC40, and IBM Blue Gene/Q), and on a heterogeneous GPU-CPU system (Cray XK7). As the bottlenecks shift from being compute-bound DGEMM's to communication-bound collectives as the size of the molecular system scales, we adopt two radically different parallelization approaches for handling load-imbalance, tasking and bulk synchronous models. Nevertheless, we preserve a unified interface to both programming models to maintain the productivity of computational quantum chemists.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2022

Ponderomotive force driven density modifications parallel to B on the LAPD

The ponderomotive force has previously been identified as a possible driver of observed density modifications close to radio frequency actuators during operation. This nonlinear force redistributes density in regions of gradients in the magnitude of an oscillating electric field and describes the influence of the fast time scale RF wave dynamics on slow time scale plasma transport. Depletion of the saturation current (a proxy for the density) measured at the Large Plasma Device (LAPD) was 30–35% during ion cyclotron range of frequencies operation. A coupled 1D plasma transport and cold plasma frequency domain wave solver was developed to self-consistently describe ponderomotive effects and was used to compare with results obtained from the LAPD experiment. The scaled current density driver for the wave model yielded an RF B field in close agreement with two components of the experimental data. However, the 1D parallel model did not accurately reproduce the amplitude or spatial distribution observed in experimental measurements of By. Within the limitations of the 1D model, initial simulation results showed that the ponderomotive force depleted up to 8% for high power (1 MW) and around 1% for the experimental power of 120 kW. This could suggest that the ponderomotive force is not the main driver of density modification for the LAPD experiments presented in this paper. Higher fidelity tools of at least 2D will be required to give a more realistic description of the RF E fields and the effect of the ponderomotive force on the LAPD.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗
DOE OSTI2026

Accelerating Bilevel Optimization With Hierarchical Many-Threaded Parallel Differential Evolution

Bilevel optimization is encountered in many relevant real-world applications. The main feature of this type of problem is that an upper-level optimization problem is constrained by a nested lower-level optimization problem. Because of this nested structure, bilevel problems (BLPs) are usually computationally expensive to solve. Differential evolution (DE) has demonstrated promising results in solving BLPs of relatively small scales. As the problem scale increases, the decision space becomes intrinsically larger, requiring a growing number of function evaluations for the method to work properly. In this context, heavy parallelization and high-performance computing techniques are indispensable to enable the resolution of more complex and challenging optimization problems. Hence, we propose a hierarchical many-threaded parallel DE approach for BLPs, where both levels are parallelized. The computational experiments demonstrate that the parallel implementation achieved runtime speeds ranging from 44 to 2559 times faster than the sequential version on a well-known scalable SMD benchmark test problem when executed on an NVIDIA A100 GPU. The findings indicate that the algorithm’s convergence is strongly influenced by the number of both upper- and lower-level generations. Moreover, the success of experiments with large-scale problems is closely linked to the choice of small population sizes.

Dufek, Amanda S↗
DOE OSTI2017

Toward General Software Level Silent Data Corruption Detection for Parallel Applications

Silent data corruption (SDC) poses a great challenge for high-performance computing (HPC) applications as we move to extreme-scale systems. Mechanisms have been proposed that are able to detect SDC in HPC applications by using the peculiarities of the data (more specifically, its “smoothness” in time and space) to make predictions. However, these data-analytic solutions are still far from fully protecting applications to a level comparable with more expensive solutions such as full replication. In this work, we propose partial replication to overcome this limitation. More specifically, we have observed that not all processes of an MPI application experience the same level of data variability at exactly the same time. Thus, we can smartly choose and replicate only those processes for which the lightweight data-analytic detectors would perform poorly. In addition, we propose a new evaluation method based on the probability that a corruption will pass unnoticed by a particular detector (instead of just reporting overall single-bit precision and recall). Here in our experiments, we use four applications dealing with different explosions. Finally, our results indicate that our new approach can protect the MPI applications analyzed with 7–70% less overhead (depending on the application) than that of full duplication with similar detection recall.

97 MATHEMATICS AND COMPUTING↗