Search NASA⌕ Search

SEARCH · Search NASA

Results for “threads”

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 253 records · Page 14

Spray: Sparse Reductions of Arrays in OPEN MP

We present SPRAY, an open-source header-only C++ library for sparse reductions of arrays. SPRAY is meant for applications in which a large array is collaboratively updated by multiple threads using an associative and commutative operation such as +=. Especially when each thread accesses only parts of the array, SPRAY can perform significantly better than OPENMP's built-in reduction clause or atomic updates, while also using less memory than the former. SPRAY provides both an easy-to-use interface that can serve as a drop-in replacement for OPENMP reductions and a selection of reducer objects that accumulate the final result in different thread-safe ways. We demonstrate SPRAY through multiple test cases including the LULESH shock hydrodynamics code and a transpose-matrix-vector multiplication for sparse matrices stored in CSR format. SPRAY reductions outperform built-in OPENMP reductions consistently, in some cases improving run time and memory overhead by 20X, and even beating domain-specific approaches such as Intel MKI, by over 2X in some cases. Furthermore, SPRAY reductions have a minimal impact on the code base, requiring only a few lines of source code changes. Once in place, SPRAY reduction schemes can be switched easily, allowing performance portability and tuning opportunities by separating performance-critical implementation details from application code.

OpenMP↗

Accelerating Flash-X Simulations with Asynchronous I/O

Most high-fidelity physics simulation codes, such as Flash-X, need to save intermediate results (checkpoint files) to restart or gain insights into the evolution of the simulation. These simulation codes save such intermediate files synchronously, where computation is stalled while the data is written to storage. Depending on the problem size and computational requirements, this file write time can be a substantial portion of the total simulation time. In order to hide the I/O latency of checkpointing, asynchronous I/O methods have been introduced. These methods use background threads for performing I/O while the main threads continue with the simulation. The usage of background threads can compete for resources on the node as well as with communication. In this paper, we evaluate the overheads and the overall benefit of asynchronous I/O in HDF5 to simulations. Results from real-world high-fidelity simulations on the Summit supercomputer show that I/O operation is overlapped with application communication or computation or both, effectively hiding some or all of the I/O latency. Our evaluation shows that while using asynchronous I/O adds overhead to the application, the I/O time reduction is more significant, resulting in overall up to 1.5X performance speedup.

Jain, Rajeev↗

Fast GPU-Based Generation of Large Graph Networks From Degree Distributions

Synthetically generated, large graph networks serve as useful proxies to real-world networks for many graph-based applications. The ability to generate such networks helps overcome several limitations of real-world networks regarding their number, availability, and access. Here, we present the design, implementation, and performance study of a novel network generator that can produce very large graph networks conforming to any desired degree distribution. The generator is designed and implemented for efficient execution on modern graphics processing units (GPUs). Given an array of desired vertex degrees and number of vertices for each desired degree, our algorithm generates the edges of a random graph that satisfies the input degree distribution. Multiple runtime variants are implemented and tested: 1) a uniform static work assignment using a fixed thread launch scheme, 2) a load-balanced static work assignment also with fixed thread launch but with cost-aware task-to-thread mapping, and 3) a dynamic scheme with multiple GPU kernels asynchronously launched from the CPU. The generation is tested on a range of popular networks such as Twitter and Facebook, representing different scales and skews in degree distributions. Results show that, using our algorithm on a single modern GPU (NVIDIA Volta V100), it is possible to generate large-scale graph networks at rates exceeding 50 billion edges per second for a 69 billion-edge network. GPU profiling confirms high utilization and low branching divergence of our implementation from small to large network sizes. For networks with scattered distributions, we provide a coarsening method that further increases the GPU-based generation speed by up to a factor of 4 on tested input networks with over 45 billion edges.

97 MATHEMATICS AND COMPUTING↗

A torque, tension and stress corrosion evaluation of high strength A286 bolts

The problems associated with overtorque applied to the Booster Separation Motor (BSM) Igniter Adapter high strength 200 KSI (1379 Mpa) A286 CRES bolts and the threaded holes of the 7075-T73 aluminum alloy BSM cases are addressed. The evaluation included torque, tensile, and stress corrosion tests incorporating the A286 CRES bolts and the 7075-T73 aluminum alloy BSM cases. The tensile test data includes ultimate tensile load (UTL), Johnson's 2/3 yield load (J2/3YL), proportional limit load (PLL), and total bolt stretch. Torque tension data includes torque, torque induced load, and positive and negative break-away torque. Stress corrosion test data reflect the overtorque and the resulting torque induced loads sustained by the A286 CRES bolts torqued into a 7075-T73 aluminum alloy forged dome with threaded holes. After 60 days of salt fog exposure, the positive and the negative break-away torques, the subsequent mechanical property tensile test results, and the BSM dome threaded hole axial tensile pullout loads are reported.

Montano, J. W.↗

Self-Aligning Coupler

Joint reduces assembly time and eliminates fumbling. Self-aligning coupler easy to use for people wearing heavy gloves or other restrictive clothing. Consists of two threaded sections, one with blade, other with slot - joined by threaded collar. Blade fits precisely in slot. Notch in blade engages pin in slot to form temporary attachment. Collar turned on continuous thread of joined sections to form tight, rigid joint. Designed for assembly of structures by astronauts in space suits, coupler used on Earth by firefighters wearing protective garments, technicians handling hazardous materials, and others working underwater or in other difficult environments.

Cooney, Earl T.↗

Evaluation of newly formulated Dow Corning 321 dry film lubricant

An evaluation of the newly formulated Dow Corning 321 dry film lubricant was performed. The purpose of the evaluation was to compare lubricating characteristics of Dow Corning 321 (STW4-2955, SCN No. 3) to those of Molykote 321R (STW4-2955). Ten igniter bolts were installed and torqued on test plates using the old formulation thread lubricant (Molykote 321R), and 10 bolts were installed using the new formulation (Dow Corning 321). After bolt removal, no signs of galling were found on any of the bolts or test plates threaded holes. Average torque-load values for each formulation were very close. Test results showed there are no significant differences in lubrication abilities between Molykote 321R and Dow Corning 321. It is recommended that, once current supplies of Molykote 321R are depleted, Dow Corning 321 dry film lubricant be used in place of Molykote 321R as a thread lubricant on redesigned solid rocket motor assemblies.

Cook, M.↗

Lunar drill footplate and casing

To prevent hole collapse during lunar drilling operations, a casing has been devised of a graphite reinforced polyimide composite which will be able to withstand the lunar environment. Additionally, this casing will be inserted into the ground in segments two meters long which will penetrate the regolith simultaneously with the auger. The vertical action of the mobile platform will provide a downward force to the casing string through a special adaptor, giving the casing the needed impetus to sink the anticipated depth of ten meters. Casing segments will be connected with a simple snap arrangement. Excess casing will be cut off by a cylindrical cutting tool which will also transport the excess casing away from the hole. A footplate will be incorporated to grasp the auger rod string during rod segment additions or removals. The footplate grasping mechanism will consist of a set of vice-like arms, one end of each bearing threaded to a common power screw. The power screw will be threaded such that one end's thread pitch opposes that of the other end. The weight of the auger and rod string will be transmitted through the arms to the power screw and absorbed by a set of three ball bearing assemblies. The power screw will be driven by a one-half horsepower brushless motor actuated by radio control. The footplate will rest on four short legs and be anchored with pins that are an integral part of each leg.

Maassen, Erik C.↗

Cantilever clamp fitting

A device is disclosed for sealing and clamping a cylindrical element which is to be attached to an object such as a wall, a pressurized vessel or another cylindrical element. The device includes a gland having an inner cylindrical wall, which is threaded at one end and is attached at a bendable end to a deformable portion, which in turn is attached to one end of a conical cantilever structure. The other end of the cantilever structure connects at a bendable area to one end of an outer cylindrical wall. The opposite end of cylindrical wall terminates in a thickened portion, the radially outer surface of which is adapted to accommodate a tool for rotating the gland. The terminal end of cylindrical wall also includes an abutment surface, which is adapted to engage a seal, which in turn engages a surface of a receiver. The receiver further includes a threaded portion for engagement with the threaded portion of gland whereby a tightening rotation of gland relative to receiver will cause relative movement between cylindrical walls and of gland. This movement causes a rotation of the conical structure and thus a bending action at bending area and at the bending end of the upper end of inner cylindrical wall. These rotational and bending actions result in a forcing of the deformable portion radially inwardly so as to contact and deform a pipe. This forcible contact creates a seal between gland and pipe, and simultaneously clamps the pipe in position.

Melton, Patrick B.↗

Quick application/release nut with engagement indicator (commercial application of an innovative nut design)

This is an assembly which permits a fastener to be inserted or removed from either side with an indicator of fastener engagement. The nut has a plurality of segments, preferably at least three segments, which are internally threaded, spring loaded apart by an internal spring, and has detents on opposite sides which force the nut segments into operative engagement with a threaded member when pushed in and release the segments for quick insertion or removal of the fastener when moved out. When the nut is installed, end pressure on the detents presses the nut segments into operative engagement with a threaded member where continued rotation locks the structure together with the detents depressed to indicate positive locking engagement of the nut. On removal, counterclockwise rotation relieves the endwise pressure on the detents, permitting internal springs to force the detents outward, allowing the nut segments to move outward and separate to permit quick removal of the fastener.

Wright, Jay M.↗

Quick application/release nut with engagement indicator

A composite nut is shown which permits a fastener to be inserted or removed from either side with an indicator of fastener engagement. The nut has a plurality of segments, preferably at least three segments, which are internally threaded, spring loaded apart by an internal spring, and has detents on opposite sides which force the nut segments into operative engagements with a threaded member when pushed in and release the segments for quick insertion or removal of the nut when moved out. When the nut is installed, end pressure on one of the detents presses the nut segments into operative engagement with a threaded member where continued rotation locks the structure together with the detents depressed to indicate positive locking engagement of the nut. On removal, counterclockwise rotation of the nut relieves the endwise pressure on the detents, permitting internal springs to force the detents outward and allowing the nut segments to move outward and separate to permit quick removal of the fastener.

Wright, Jay M.↗

A designed experiment in stitched/RTM composites

The damage tolerance of composite laminates can be significantly improved by the addition of through-the-thickness fibrous reinforcement such as stitching. However, there are numerous stitching parameters which can be independently varied, and their separate and combined effects on mechanical properties need to be determined. A statistically designed experiment (a 2(sup 5-1) fractional factorial, also known as a Taguchi L16 test matrix) used to evaluate five important parameters is described. The effects and interactions of stitch thread material, stitch thread strength, stitch row spacing and stitch pitch are examined for both thick (48 ply) and thin (16 ply) carbon/epoxy (AS4/E905L) composites. Tension, compression and compression after impact tests are described. Preliminary results of completed tension testing are discussed. Larger threads decreased tensile strength. Panel thickness was found not to be an important stitching parameter for tensile properties. Tensile modulus was unaffected by stitching.

Dickinson, Larry C.↗

Adjustable-Length Strut Withstands Large Cyclic Loads

Adjustable-length strut designed for installation in structure subjected to large cyclic loads. Partly resembles large turnbuckle: includes oppositely threaded eyebolts engaging correspondingly threaded holes at opposite ends of shaft, and shaft turned to adjust length. However, unlike in turnbuckle, length setting not fixed by use of simple jam nuts: instead, length setting fixed by use of more complex threaded-end flanges partly resembling jam nuts but function somewhat differently.

Carner, Fred P.↗

Tubular Coupling

A system for coupling a vascular overflow graft or cannula to a heart pump. A pump pipe outlet is provided with an external tapered surface which receives the end of a compressible connula. An annular compression ring with a tapered internal bore surface is arranged about the cannula with the tapered internal surface in a facing relationship to the external tapered surface. The angle of inclination of the tapered surfaces is converging such that the spacing between the tapered surfaces decreases from one end of the external tapered surface to the other end thereby providing a clamping action of the tapered surface on a cannula which increases as a function of the length of cannula segment between the tapered surfaces. The annular compression ring is disposed within a tubular locking nut which threadedly couples to the pump and provides a compression force for urging the annular ring onto the cannula between the tapered surfaces. The nut has a threaded connection to the pump body. The threaded coupling to the pump body provides a compression force for the annular ring. The annular ring has an annular enclosure space in which excess cannula material from the compression between the tapered surfaces to "bunch up" in the space and serve as an enlarged annular ring segment to assist holding the cannula in place. The clamped cannula provides a seamless joint connection to the pump pipe outlet where the clamping force is uniformly applied to the cannula because of self alignment of the tapered surfaces. The nut can be easily disconnected to replace the pump if necessary.

Rosenbaum, Bernard J.↗

Efficient Parallelization of a Dynamic Unstructured Application on the Tera MTA

The success of parallel computing in solving real-life computationally-intensive problems relies on their efficient mapping and execution on large-scale multiprocessor architectures. Many important applications are both unstructured and dynamic in nature, making their efficient parallel implementation a daunting task. This paper presents the parallelization of a dynamic unstructured mesh adaptation algorithm using three popular programming paradigms on three leading supercomputers. We examine an MPI message-passing implementation on the Cray T3E and the SGI Origin2OOO, a shared-memory implementation using cache coherent nonuniform memory access (CC-NUMA) of the Origin2OOO, and a multi-threaded version on the newly-released Tera Multi-threaded Architecture (MTA). We compare several critical factors of this parallel code development, including runtime, scalability, programmability, and memory overhead. Our overall results demonstrate that multi-threaded systems offer tremendous potential for quickly and efficiently solving some of the most challenging real-life problems on parallel computers.

Oliker, Leonid↗

Issues Related to Cleaning Complex Geometry Surfaces with ODC-Free Solvents

Implementing ozone depleting chemicals (ODC)-free solvents into full-scale reusable solid rocket motor cleaning operations has presented problems due to the low vapor pressures of the solvents. Because of slow evaporation, solvent retention is a problem on porous substrates or on surfaces with irregular geometry, such as threaded boltholes, leak check ports, and nozzle backfill joints. The new solvents are being evaluated to replace 1,1,1-trichloroethane, which readily evaporates from these surfaces. Selection of the solvents to be evaluated on full-scale hardware was made based on results of subscale tests performed with flat surface coupons, which did not manifest the problem. Test efforts have been undertaken to address concerns with the slow-evaporating solvents. These concerns include effects on materials due to long-term exposure to solvent, potential migration from bolthole threads to seal surfaces, and effects on bolt loading due to solvent retention in threads. Tests performed to date have verified that retained solvent does not affect materials or hardware performance. Process modifications have also been developed to assist drying, and these can be implemented if additional drying becomes necessary.

Bradford, Blake F.↗

Optical and Probe Diagnostics Applied to Reacting Flows

We plan to explore three major threads during the fellowship period. The first interrogates the flame synthesis of carbon nanotubes using aerosol catalysts. Laser light scattering will reveal changes in particle size at various heights above the burner. Analysis of the flame gas by mass spectroscopy will reveal the chemical composition of the mixture. Finally, absorption measurements will map the nanotube concentration within the flow. The second thread explores soot oxidation kinetics. Cavity ring-down absorption measurements of the carbonaceous aerosol can provide a measure of the mass concentration with time and, hence, an oxidation rate. Spectroscopic and direct probe measurements will provide the temperature of the system needed for subsequent modeling. The third thread will explore the details of turbulent flame dynamics. Laser induced incandescence will be applied to measurements of soot volume fraction in a 2-d configuration. Analysis of seed tracer particles by planar laser light MIE scattering will reveal the elemental fuel mixture fraction in the flames. Cavity ring-down spectroscopy, a pulsed transient absorption method, will determine the instantaneous mass loading and its fluctuation. Finally, fluorescence measurements will investigate the formation of PAH's in these flames.

Ticich, Thomas M.↗

Applying Jlint to Space Exploration Software

Java is a very successful programming language which is also becoming widespread in embedded systems, where software correctness is critical. Jlint is a simple but highly efficient static analyzer that checks a Java program for several common errors, such as null pointer exceptions, and overflow errors. It also includes checks for multi-threading problems, such as deadlocks and data races. The case study described here shows the effectiveness of Jlint in find-false positives in the multi-threading warnings gives an insight into design patterns commonly used in multi-threaded code. The results show that a few analysis techniques are sufficient to avoid almost all false positives. These techniques include investigating all possible callers and a few code idioms. Verifying the correct application of these patterns is still crucial, because their correct usage is not trivial.

Artho, Cyrille↗

Life's origin: the cosmic, planetary and biological processes

From elements formed in interstellar furnaces to humans peering back at the stars, the evolution of life has been a long, intricate and perhaps inevitable process. Life as we know it requires a planet orbiting a star at just the right distance so that water can exist in liquid form. It needs a rich supply of chemicals and energy sources. On Earth, the combination of chemistry and energy generated molecules that evolved ways of replicating themselves and of passing information from one generation to the next. Thus, the thread of life began. This chart traces the thread, maintained by DNA molecules for much of its history, as it weaves its way through the primitive oceans, gaining strength and diversity along the way. Organisms eventually moved onto the land, where advanced forms, including humans, ultimately arose. Finally, assisted by a technology of its own making, life has reached back out into space to understand its own origins, to expand into new realms, and to seek other living threads in the cosmos.

NASA Center ARC↗