Search NASA⌕ Search

SEARCH · Search NASA

Results for “Threading”

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 379 records · Page 21

Dislocation half-loop control for optimal V-defect density in GaN-based light emitting diodes

V-defects are morphological defects that typically form on threading dislocations during epitaxial growth of $(0001)$-oriented GaN layers. A V-defect is a hexagonal pyramid-shaped depression with six {$10\bar{1}1$}-oriented sidewalls. These semipolar sidewalls have a lower polarization barrier than the polarization barriers present between the polar c-plane quantum wells and quantum barriers and can laterally inject carriers directly into quantum wells in GaN-based light emitting diodes (LEDs). This is especially important, as the high polarization field in c-plane GaN is a significant factor in the high forward voltage of GaN LEDs. The optimal V-defect density for efficient lateral carrier injection in a GaN LED (∼10 9 cm −2 ) is typically an order of magnitude higher than the threading dislocation density of GaN grown on patterned sapphire substrates (∼10 8 cm −2 ). Pure-edge dislocation loops have been known to exist in GaN, and their formation into large V-defects via low-temperature growth with high Si-doping has recently been studied. Here, in this work, we develop a method for pure-edge threading dislocation half-loop formation and density control via disilane flow, growth temperature, and thickness of the half-loop generation layer. We also develop a method of forming the threading dislocation half-loops into V-defects of comparable size to those originating from substrate threading dislocations.

Crystallographic defects↗

TANGLED MAGNETIC FIELDS IN SOLAR PROMINENCES

Solar prominences are an important tool for studying the structure and evolution of the coronal magnetic field. Here we consider so-called hedgerow prominences, which consist of thin vertical threads. We explore the possibility that such prominences are supported by tangled magnetic fields. A variety of different approaches are used. First, the dynamics of plasma within a tangled field is considered. We find that the contorted shape of the flux tubes significantly reduces the flow velocity compared to the supersonic free fall that would occur in a straight vertical tube. Second, linear force-free models of tangled fields are developed, and the elastic response of such fields to gravitational forces is considered. We demonstrate that the prominence plasma can be supported by the magnetic pressure of a tangled field that pervades not only the observed dense threads but also their local surroundings. Tangled fields with field strengths of about 10 G are able to support prominence threads with observed hydrogen density of the order of 10{sup 11} cm{sup -3}. Finally, we suggest that the observed vertical threads are the result of Rayleigh-Taylor instability. Simulations of the density distribution within a prominence thread indicate that the peak density is much larger than the average density. We conclude that tangled fields provide a viable mechanism for magnetic support of hedgerow prominences.

79 ASTRONOMY AND ASTROPHYSICS↗

Observations and implications of large-amplitude longitudinal oscillations in a solar filament

On 2010 August 20, an energetic disturbance triggered large-amplitude longitudinal oscillations in a nearby filament. The triggering mechanism appears to be episodic jets connecting the energetic event with the filament threads. In the present work, we analyze this periodic motion in a large fraction of the filament to characterize the underlying physics of the oscillation as well as the filament properties. The results support our previous theoretical conclusions that the restoring force of large-amplitude longitudinal oscillations is solar gravity, and the damping mechanism is the ongoing accumulation of mass onto the oscillating threads. Based on our previous work, we used the fitted parameters to determine the magnitude and radius of curvature of the dipped magnetic field along the filament, as well as the mass accretion rate onto the filament threads. These derived properties are nearly uniform along the filament, indicating a remarkable degree of cohesiveness throughout the filament channel. Moreover, the estimated mass accretion rate implies that the footpoint heating responsible for the thread formation, according to the thermal nonequilibrium model, agrees with previous coronal heating estimates. We estimate the magnitude of the energy released in the nearby event by studying the dynamic response of the filament threads, and discuss the implications of our study for filament structure and heating.

79 ASTRONOMY AND ASTROPHYSICS↗

LARGE-AMPLITUDE LONGITUDINAL OSCILLATIONS IN A SOLAR FILAMENT

We have developed the first self-consistent model for the observed large-amplitude oscillations along filament axes that explains the restoring force and damping mechanism. We have investigated the oscillations of multiple threads formed in long, dipped flux tubes through the thermal nonequilibrium process, and found that the oscillation properties predicted by our simulations agree with the observed behavior. We then constructed a model for the large-amplitude longitudinal oscillations that demonstrates that the restoring force is the projected gravity in the tube where the threads oscillate. Although the period is independent of the tube length and the constantly growing mass, the motions are strongly damped by the steady accretion of mass onto the threads by thermal nonequilibrium. The observations and our model suggest that a nearby impulsive event drives the existing prominence threads along their supporting tubes, away from the heating deposition site, without destroying them. The subsequent oscillations occur because the displaced threads reside in magnetic concavities with large radii of curvature. Our model yields a powerful seismological method for constraining the coronal magnetic field and radius of curvature of dips. Furthermore, these results indicate that the magnetic structure is most consistent with the sheared-arcade model for filament channels.

79 ASTRONOMY AND ASTROPHYSICS↗

TIME EVOLUTION OF PLASMA PARAMETERS DURING THE RISE OF A SOLAR PROMINENCE INSTABILITY

We present high-spatial resolution spectropolarimetric observations of a quiescent hedgerow prominence taken in the He I 1083.0 nm triplet. The observation consisted of a time series in sit-and-stare mode of ∼36 minutes duration. The spectrograph's slit crossed the prominence body and we recorded the time evolution of individual vertical threads. Eventually, we observed the development of a dark Rayleigh-Taylor plume that propagated upward with a velocity, projected onto the plane of the sky, of 17 km s{sup –1}. Interestingly, the plume apex collided with the prominence threads pushing them aside. We inferred Doppler shifts, Doppler widths, and magnetic field strength variations by interpreting the He I Stokes profiles with the HAZEL code. The Doppler shifts show that clusters of threads move coherently while individual threads have oscillatory patterns. Regarding the plume we found strong redshifts (∼9-12 km s{sup –1}) and large Doppler widths (∼10 km s{sup –1}) at the plume apex when it passed through the prominence body and before it disintegrated. We associate the redshifts with perspective effects while the Doppler widths are more likely due to an increase in the local temperature. No local variations of the magnetic field strength associated with the passage of the plume were found; this leads us to conclude that the plumes are no more magnetized than the surroundings. Finally, we found that some of the threads' oscillations are locally damped, what allowed us to apply prominence seismology techniques to infer additional prominence physical parameters.

79 ASTRONOMY AND ASTROPHYSICS↗

Characterization of the torque limits and clamping force relationships for small stainless steel screws in tensile loaded joints of various metals

This study originated during the design of ChemCam, a Laser Induced Breakdown Spectroscopy (LIBS) and imaging instrument being developed for NASA's Mars Science Lab Rover. The mission needs for miniaturization, reduced weight, high reliability, minimal use of thread locking compounds, and the ability to handle harsh environmental conditions dictated the use of small, high strength screws to be threaded into a variety of metal alloys including Be-S200f, Al-6061-T6, Mg-ZK60A-T5, and Ti-6Al-4V The lack of a credible fastener torque database for small (No.0 through No.8) high strength stainless steel screws in various parent materials, led to the development of an experimental program to characterize the following: (A) The screw torque value versus angular rotation (which indicates yielding in the screw or parent material) as a function of screw diameter, screw head configuration, depth of thread engagement, type of parent material, type of surface treatment on parent material, presence of thread locking compound, repeatable threaded hole use, and degree of screw pedigree. (B) The relationship between fastener torque and clamping force for a subset of the above mentioned variables. The database generated from this study will serve as a design reference for utilizing small stainless steel fasteners and provide trending information for other researchers who may be interested in broadening its range of parameters. This paper reviews the related fastener torque and clamping force information from the literature, describes the experimental screw torque and clamping force monitoring equipment, presents the test matrix and experimental procedures, and discusses the empirical results.

36 MATERIALS SCIENCE↗

Improving concurrency and asynchrony in multithreaded MPI applications using software offloading

We present a new approach for multithreaded communication and asynchronous progress in MPI applications, wherein we offload communication processing to a dedicated thread. The central premise is that given the rapidly increasing core counts on modern systems, the improvements in MPI performance arising from dedicating a thread to drive communication outweigh the small loss of resources for application computation, particularly when overlap of communication and computation can be exploited. Our approach allows application threads to make MPI calls concurrently, enqueuing these as communication tasks to be processed by a dedicated communication thread. This not only guarantees progress for such communication operations, but also reduces load imbalance. Our implementation additionally significantly reduces the overhead of mutual exclusion seen in existing implementations for applications using MPI THREAD MULTIPLE. Our technique requires no modification to the application, and we demonstrate significant performance improvement (up to 2X) for QCD, 1-D FFT and deep learning CNN applications.

Vaidyanathan, Karthikeyan↗

Deploy Nalu/Kokkos algorithmic infrastructure with performance benchmarking.

The former Nalu interior heterogeneous algorithm design, which was originally designed to manage matrix assembly operations over all elemental topology types, has been modified to operate over homogeneous collections of mesh entities. This newly templated kernel design allows for removal of workset variable resize operations that were formerly required at each loop over a Sierra ToolKit (STK) bucket (nominally, 512 entities in size). Extensive usage of the Standard Template Library (STL) std::vector has been removed in favor of intrinsic Kokkos memory views. In this milestone effort, the transition to Kokkos as the underlying infrastructure to support performance and portability on many-core architectures has been deployed for key matrix algorithmic kernels. A unit-test driven design effort has developed a homogeneous entity algorithm that employs a team-based thread parallelism construct. The STK Single Instruction Multiple Data (SIMD) infrastructure is used to interleave data for improved vectorization. The collective algorithm design, which allows for concurrent threading and SIMD management, has been deployed for the core low-Mach element- based algorithm. Several tests to ascertain SIMD performance on Intel KNL and Haswell architectures have been carried out. The performance test matrix includes evaluation of both low- and higher-order methods. The higher-order low-Mach methodology builds on polynomial promotion of the core low-order control volume nite element method (CVFEM). Performance testing of the Kokkos-view/SIMD design indicates low-order matrix assembly kernel speed-up ranging between two and four times depending on mesh loading and node count. Better speedups are observed for higher-order meshes (currently only P=2 has been tested) especially on KNL. The increased workload per element on higher-order meshes bene ts from the wide SIMD width on KNL machines. Combining multiple threads with SIMD on KNL achieves a 4.6x speedup over the baseline, with assembly timings faster than that observed on Haswell architecture. The computational workload of higher-order meshes, therefore, seems ideally suited for the many-core architecture and justi es further exploration of higher-order on NGP platforms. A Trilinos/Tpetra-based multi-threaded GMRES preconditioned by symmetric Gauss Seidel (SGS) represents the core solver infrastructure for the low-Mach advection/diffusion implicit solves. The threaded solver stack has been tested on small problems on NREL's Peregrine system using the newly developed and deployed Kokkos-view/SIMD kernels. fforts are underway to deploy the Tpetra-based solver stack on NERSC Cori system to benchmark its performance at scale on KNL machines.

97 MATHEMATICS AND COMPUTING↗

iHarmonizer: improving the disk efficiency of I/O-intensive multithreaded codes

Challenged by serious power and thermal constraints, and limited by available instruction-level parallelism, processor designs have evolved to multi-core architectures. These architectures, many augmented with native simultaneous multithreading, are driving software developers to use multithreaded programs to exploit thread-level parallelism. While multithreading is well-known to introduce concerns of data dependency and CPU load balance, less known is that the uncertainty of relative progress of thread execution can cause patterns of I/O requests, issued by different threads, to be effectively random and so significantly degrade hard-disk efficiency. This effect can severely offset the performance gains from parallel execution, especially for I/O-intensive programs. Retaining the benefits of multithreading while not reducing I/O efficiency is an urgent and challenging problem. We propose a user-level scheme, iHarmonizer, to streamline the servicing of I/O requests from multiple threads in OpenMP programs. Specifically, we use the compiler to insert code into OpenMP programs so that data usage can be transmitted at run time to a supporting run-time library; this library in turn prefetches data in a disk-friendly way and coordinates threads execution according to the availability of their requested data. Transparently to the programmer, iHarmonizer makes a multithreaded program I/O efficient while maintaining the benefits of parallelism. Our experiments show that iHarmonizer can significantly speed up the execution of a representative set of I/O-intensive scientific benchmarks.

97 MATHEMATICS AND COMPUTING↗

Implementing a Gaussian Process Learning Algorithm in Mixed Parallel Environment

In this paper, we present a scalability analysis of a parallel Gaussian process training algorithm to simultaneously analyze a massive number of time series. We study three different parallel implementations: using threads, MPI, and a hybrid implementation using threads and MPI. We compare the scalability for the multi-threaded implementation on three different hardware platforms: a Mac desktop with two quad-core Intel Xeon processors (16 virtual cores), a Linux cluster node with four quad-core 2.3 GHz AMD Opteron processors, and SGI Altix ICE 8200 cluster node with two quad-core Intel Xeon processors (16 virtual cores). We also study the scalability of the MPI based and the hybrid MPI and thread based implementations on the SGI cluster with 128 nodes (2048 cores). Experimental results show that the hybrid implementation scales better than the multi-threaded and MPI based implementations. The hybrid implementation, using 1536 cores, can analyze a remote sensing data set with over 4 million time series in nearly 5 seconds while the serial algorithm takes nearly 12 hours to process the same data set.

99 GENERAL AND MISCELLANEOUS↗

Message passing with queues and channels

In an embodiment, a send thread receives an identifier that identifies a destination node and a pointer to data. The send thread creates a first send request in response to the receipt of the identifier and the data pointer. The send thread selects a selected channel from among a plurality of channels. The selected channel comprises a selected hand-off queue and an identification of a selected message unit. Each of the channels identifies a different message unit. The selected hand-off queue is randomly accessible. If the selected hand-off queue contains an available entry, the send thread adds the first send request to the selected hand-off queue. If the selected hand-off queue does not contain an available entry, the send thread removes a second send request from the selected hand-off queue and sends the second send request to the selected message unit.

97 MATHEMATICS AND COMPUTING↗

Processor power monitoring and control with dynamic load balancing

Apparatus and methods may provide for subscribing a thread to a resource monitor through a machine specific register and subscribing the thread to a class of service through the machine specific register. The resource monitor or the class of service for the thread may be changed without interrupting the thread. The power allocated to the processor core may be changed based on the selected class of service for the thread.

97 MATHEMATICS AND COMPUTING↗

Cache utilization and eviction based on allocated priority tokens

A system, method, and computer program product are provided for providing prioritized access for multithreaded processing. The method includes the steps of allocating threads to process a workload and assigning a set of priority tokens to at least a portion of the threads. Access to a resource, by each one of the threads, is based on the priority token assigned to the thread and the threads are executed by a multithreaded processor to process the workload.

Johnson, Daniel Robert↗

Data Parallel Bin-Based Indexing for Answering Queries on Multi-Core Architectures

The multi-core trend in CPUs and general purpose graphics processing units (GPUs) offers new opportunities for the database community. The increase of cores at exponential rates is likely to affect virtually every server and client in the coming decade, and presents database management systems with a huge, compelling disruption that will radically change how processing is done. This paper presents a new parallel indexing data structure for answering queries that takes full advantage of the increasing thread-level parallelism emerging in multi-core architectures. In our approach, our Data Parallel Bin-based Index Strategy (DP-BIS) first bins the base data, and then partitions and stores the values in each bin as a separate, bin-based data cluster. In answering a query, the procedures for examining the bin numbers and the bin-based data clusters offer the maximum possible level of concurrency; each record is evaluated by a single thread and all threads are processed simultaneously in parallel. We implement and demonstrate the effectiveness of DP-BIS on two multi-core architectures: a multi-core CPU and a GPU. The concurrency afforded by DP-BIS allows us to fully utilize the thread-level parallelism provided by each architecture--for example, our GPU-based DP-BIS implementation simultaneously evaluates over 12,000 records with an equivalent number of concurrently executing threads. In comparing DP-BIS's performance across these architectures, we show that the GPU-based DP-BIS implementation requires significantly less computation time to answer a query than the CPU-based implementation. We also demonstrate in our analysis that DP-BIS provides better overall performance than the commonly utilized CPU and GPU-based projection index. Finally, due to data encoding, we show that DP-BIS accesses significantly smaller amounts of data than index strategies that operate solely on a column's base data; this smaller data footprint is critical for parallel processors that possess limited memory resources (e.g., GPUs).

97 MATHEMATICS AND COMPUTING↗

Implementing and Evaluating Multithreaded Triad Census Algorithms on the Cray XMT

Commonly represented as directed graphs, social networks depict relationships and behaviors among social entities such as people, groups, and organizations. Social network analysis denotes a class of mathematical and statistical methods designed to study and measure social networks. Beyond sociolo-gy, social network analysis methods are being applied to other types of data in other domains such as bioinformatics, computer networks, national security, and economics. For particular problems, the size of a social network can grow to millions of nodes and tens of millions of edges or more. In such cases, researchers could benefit from the application of social network analysis algorithms on high-performance architectures and systems. The Cray XMT is a third generation multithreaded system based on the Cray XT-3/4 platform. Like most other multithreaded architectures, the Cray XMT is designed to tolerate memory access latencies by switching context between threads. The processors maintain multiple threads of execution and util-ize hardware-based context switching to overlap the memory latency incurred by any thread with the computations from other threads. Due to its memory latency tolerance, the Cray XMT has the poten-tial of significantly improving the execution speed of irregular data-intensive applications such as those found in social network analysis. In this paper, we describe our experiences in developing and optimizing three implementations of a social network analysis method known as triadic analysis to execute on the Cray XMT. The three im-plementations possess different execution complexities, qualities, and characteristics. We evaluate how the various attributes of the codes affect their performance on the Cray XMT. We also explore the effects of different compiler options and execution strategies on the different triadic analysis im-plementations and identify general XMT programming issues and lessons learned.

45 MILITARY TECHNOLOGY, WEAPONRY, AND NATIONAL DEF↗

Power/Performance Trade-offs of Small Batched LU Based Solvers on GPUs

In this paper we propose and analyze a set of batched linear solvers for small matrices on Graphic Processing Units (GPUs), evaluating the various alternatives depending on the size of the systems to solve. We discuss three different solutions that operate with different level of parallelization and GPU features. The first, exploiting the CUBLAS library, manages matrices of size up to 32x32 and employs Warp level (one matrix, one Warp) parallelism and shared memory. The second works at Thread-block level parallelism (one matrix, one Thread-block), still exploiting shared memory but managing matrices up to 76x76. The third is Thread level parallel (one matrix, one thread) and can reach sizes up to 128x128, but it does not exploit shared memory and only relies on the high memory bandwidth of the GPU. The first and second solution only support partial pivoting, the third one easily supports partial and full pivoting, making it attractive to problems that require greater numerical stability. We analyze the trade-offs in terms of performance and power consumption as function of the size of the linear systems that are simultaneously solved. We execute the three implementations on a Tesla M2090 (Fermi) and on a Tesla K20 (Kepler).

GPGPU↗

Joining AA7099 to Ni-Cr-Mo Steel Using Single Pass Friction Stir Dovetailing and AA6061 Butter Layer

A novel single pass friction stir dovetailing (FSD) was developed to join thick plate AA7099 to Ni-Cr-Mo steel in a lap configuration through the use of an AA6061 butter layer. The butter layer approach can be shown to significantly reduce interfacial failures by inhibiting the formation of brittle Fe–Al intermetallic compounds (IMCs) which are typically formed during conventional friction stir welding (FSW) of AA7099 directly to steel. This study evaluates the strength and microstructural evolution of the aluminum/steel joint interface between single pass FSD of AA7099/AA6061(butter)/steel fabricated using three different FSD tool configurations. The butter layer was achieved by placement of a free-floating strip of AA6061 within the dovetail groove that was pre-machined in the steel substrate. Friction stir welding was then used to simultaneously join the thick plate AA7099, the AA6061 strip and the AA7099 in a single weld pass. To optimize the joining process, three different pin geometries (full thread, half smooth/half thread, no thread) were evaluated; it was demonstrated that the half smooth/half thread tool produced the strongest weld joint due to the absence of a Zn rich layer at the Al/steel interface. The efficacy of this unique processing technique was tested using lap shear tensile testing as well as high resolution transmission electron microscopy (TEM). A discussion of the new FSD technique, joint configurations, and process parameters are provided along with joint microstructural analyses and mechanical performance.

Olszta, Matthew J.↗

X-ray topography characterization of gallium nitride substrates for power device development

Gallium nitride substrates grown by the hydride vapor phase epitaxy (HVPE) method using a patterned growth process have been characterized by synchrotron monochromatic beam X-ray topography in the grazing incidence geometry. Images reveal a starkly heterogeneous distribution of dislocations with areas as large as 0 3 mm 2 containing threading dislocation densities below 10 3 cm -2 in between a grid of strain centers with higher threading dislocation densities (> 10 4 cm -2 ). Basal plane dislocation densities in these areas are as low as 104 cm -2 . By comparing the recorded images of dislocations with ray tracing simulations of expected dislocations in GaN, the Burgers vectors of the dislocations have been determined. The distribution of threading screw/mixed dislocations (TSDs/TMDs), threading edge dislocations (TEDs) and basal plane dislocations (BPDs) is discussed with implications for fabrication of power devices.

36 MATERIALS SCIENCE↗