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 145 records · Page 8

Threaded Operations on Sparse Matrices

We investigate the use of sparse matrices and OpenMP multi-threading on linear algebra operations involving them. Several sparse matrix data structures are presented. Implementation of the multi- threading primarily occurs in the level one and two BLAS functions used within the four algorithms investigated{the Power Method, Conjugate Gradient, Biconjugate Gradient, and Jacobi's Method. The bene ts of launching threads once per high level algorithm are explored.

97 MATHEMATICS AND COMPUTING↗

Exploiting Thread Parallelism for Ocean Modeling on Cray XC Supercomputers

The incorporation of increasing core counts in modern processors used to build state-of-the-art supercomputers is driving application development towards exploitation of thread parallelism, in addition to distributed memory parallelism, with the goal of delivering efficient high-performance codes. In this work we describe the exploitation of threading and our experiences with it with respect to a real-world ocean modeling application code, MPAS-Ocean. We present detailed performance analysis and comparisons of various approaches and configurations for threading on the Cray XC series supercomputers.

97 MATHEMATICS AND COMPUTING↗

Heterogeneous graphics processing unit for scheduling thread groups for execution on variable width SIMD units

A compute unit configured to execute multiple threads in parallel is presented. The compute unit includes one or more single instruction multiple data (SIMD) units and a fetch and decode logic. The SIMD units have differing numbers of arithmetic logic units (ALUs), such that each SIMD unit can execute a different number of threads. The fetch and decode logic is in communication with each of the SIMD units, and is configured to assign the threads to the SIMD units for execution based on such differing numbers of ALUs.

97 MATHEMATICS AND COMPUTING↗

The Role of the Digital Thread for Security, Resilience, and Adaptability in Manufacturing

The manufacturing sector is dramatically evolving with recent technical and digital advances. Among these, the digital thread is revolutionizing manufacturing operations well beyond the historic downloading of programs to computer numerically controlled (CNC) machine tools and the uploading of edited programs from CNC controllers. The digital thread is “the communication framework that enables a connected data flow and integrated view of the asset’s data throughout its lifecycle across traditionally siloed functional perspectives” (Leiva 2016). It is pervasive around the world and has changed the way society operates; for example, map apps are used to provide time-optimal directions to destinations using real-time traffic feedback. The digital thread makes it possible to digitally verify products, ensure that the latest technologies are deployed across the entire manufacturing ecosystem, and strengthen the workforce by making each individual more efficient and effective. These three foundational advanced manufacturing concepts will ensure a next-generation secure, resilient, and adaptable manufacturing ecosystem and sustainably support current and future needs of society (Lynn et al. 2020).

99 GENERAL AND MISCELLANEOUS↗

T RI M E ++: Multi-threaded triangular meshing in two dimensions

We present T RI M E ++, a multi-threaded software library designed for generating two-dimensional meshes for intricate geometric shapes using the Delaunay triangulation. Multi-threaded parallel computing is implemented throughout the meshing procedure, making it suitable for fast generation of large-scale meshes. Three iterative meshing algorithms are implemented: the DistMesh algorithm, the centroidal Voronoi diagram meshing, and a hybrid of the two. We compare the performance of the three meshing methods in T RI M E ++, and show that the hybrid method retains the advantages of the other two. The software library achieves significant parallel speedup when generating large-scale meshes containing between 10 4 to 10 7 points. T RI M E ++ can handle complicated geometries and generates adaptive meshes of high quality.

97 MATHEMATICS AND COMPUTING↗

Spinning the digital thread with hybrid manufacturing

Integrated process monitoring and feedback capabilities for hybrid manufacturing have shown potential as a test platform for the development and implementation of the digital thread. Additionally, hybrid manufacturing alleviates the implementation of the digital thread as fewer systems are involved for a broader base of manufacturing operations. Researchers have developed coordinated process monitoring architectures to capture and leverage various existing data streams in the hybrid manufacturing process and improve the quality of manufactured components. This work is fundamentally different because it shows how common communication structures that exist on any commercial CNC can be used to enhance CNC based manufacturing processes.

42 ENGINEERING↗

The Entropic Braiding Index ( eBI ): A Robust Metric to Account for the Diversity of Channel Scales in Multi‐Thread Rivers

Abstract The Braiding Index ( BI ), defined as the average count of intercepted channels per cross‐section, is a widely used metric for characterizing multi‐thread river systems. However, it does not account for the diversity of channels (e.g., in terms of water discharge) within different cross‐sections, omitting important information related to system complexity. Here we present a modification of BI, the Entropic Braiding Index ( eBI ), which augments the information content in BI by using Shannon Entropy to encode the diversity of channels in each cross section. eBI is interpreted as the number of “effective channels” per cross‐section, allowing a direct comparison with the traditional BI . We demonstrate the potential of the ratio BI/eBI to quantify channel disparity, differentiate types of multi‐thread systems (braided vs. anastomosed), and assess the effect of discharge variability, such as seasonal flooding, on river cross‐section stability.

54 ENVIRONMENTAL SCIENCES↗

A Locality-Based Threading Algorithm for the Configuration-Interaction Method

The Configuration Interaction (CI) method has been widely used to solve the non-relativistic many-body Schrodinger equation. One great challenge to implementing it efficiently on manycore architectures is its immense memory and data movement requirements. To address this issue, within each node, we exploit a hybrid MPI+OpenMP programming model in lieu of the traditional flat MPI programming model. Here in this paper, we develop optimizations that partition the workloads among OpenMP threads based on data locality,-which is essential in ensuring applications with complex data access patterns scale well on manycore architectures. The new algorithm scales to 256 threadson the 64-core Intel Knights Landing (KNL) manycore processor and 24 threads on dual-socket Ivy Bridge (Xeon) nodes. Compared with the original implementation, the performance has been improved by up to 7× on theKnights Landing processor and 3× on the dual-socket Ivy Bridge node.

97 MATHEMATICS AND COMPUTING↗

Thread-level parallelization and optimization of NWChem for the Intel MIC architecture

In the multicore era it was possible to exploit the increase in on-chip parallelism by simply running multiple MPI processes per chip. Unfortunately, manycore processors' greatly increased thread- and data-level parallelism coupled with a reduced memory capacity demand an altogether different approach. In this paper we explore augmenting two NWChem modules, triples correction of the CCSD(T) and Fock matrix construction, with OpenMP in order that they might run efficiently on future manycore architectures. As the next NERSC machine will be a self-hosted Intel MIC (Xeon Phi) based supercomputer, we leverage an existing MIC testbed at NERSC to evaluate our experiments. In order to proxy the fact that future MIC machines will not have a host processor, we run all of our experiments in native mode. We found that while straightforward application of OpenMP to the deep loop nests associated with the tensor contractions of CCSD(T) was sufficient in attaining high performance, significant e ort was required to safely and efeciently thread the TEXAS integral package when constructing the Fock matrix. Ultimately, our new MPI+OpenMP hybrid implementations attain up to 65× better performance for the triples part of the CCSD(T) due in large part to the fact that the limited on-card memory limits the existing MPI implementation to a single process per card. Additionally, we obtain up to 1.6× better performance on Fock matrix constructions when compared with the best MPI implementations running multiple processes per card.

CCSD(T)↗

Establishing and Maintaining the Digital Thread of Additively Manufactured Materials and Applications

Additive Manufacturing (AM) and Integrated Computational Materials Engineering (ICME) are complementary enabling technologies for design and manufacturing of “fit-for-purpose” materials. Both technologies will impact rapid material design, reduction in cost- and time-to-market for new applications, and discovery and implementation of new materials. An ICME approach to design, however, requires experimentally validated material models at multiple length and time scales, an integrated framework that can connect analysis tools with one another to ensure the digital thread of an application is maintained, and the manufacturing (e.g., AM) capability to leverage processing-structure-property-performance (PSPP) relationships to achieve spatially varying material properties where desired. AM enables the implementation of the design of an optimized, spatially varying microstructure through careful selection of the processing parameters used during an additively manufactured build. In order to establish these PSPP relations, a large amount of data is necessary, and that data must be properly captured, analyzed and maintained in an information management system that can establish the required traceability between various aspects of the design process to ensure an application’s digital thread is maintained (from design to end of life). Such an information management system must be able to capture feedstock material pedigree, resulting microstructure from various build parameters, subsequent mechanical properties derived from testing, developed material models, and enable spatial variations in material assignment in an engineering application. Furthermore, the information management system should be easily integrated with traditionally engineered materials in a single, centralized platform to enable an ICME optimization tool to explore both types of manufacturing processes. At NASA GRC, a robust, 21st century materials information management system has been previously developed with a focus towards enabling ICME. In this work, GRC’s ICME schema is extended to accommodate additively manufactured materials, enabling storage of both traditionally and additively manufactured materials in the same construct. The methodology for properly capturing additively manufactured materials across the entire material lifecycle is presented, following the previously established database best practices, as a potential framework for establishing PSPP relationships for additively manufactured materials and applying them to engineering applications.

Data management↗

Threaded Multi-Core GEMM with MoA and Cache-Blocking: Preprint

A threaded multi-core implementation of the high performance dense linear algebra matrix-matrix multiply GEMM kernel is described. This kernel is widely implemented by vendors in the basic linear algebra subroutine BLAS library. The mathematics of arrays (MoA) paradigm due to Mullin (1988) results in contiguous memory accesses by employing outer-product forms. Our performance studies demonstrate that the MoA implementation of double precision DGEMM combined with optimal cache-blocking strategies results in at least a 25% performance gain on the Intel Xeon Skylake processor over the vendor supplied Intel MKL basic linear algebra libraries. Results are presented for the NREL Eagle supercomputer. The multi-core DGEMM achieves over 100 GigaFlops/sec with eight openMP threads.

cache-blocking↗

Towards a bit threads derivation of holographic entanglement of purification

We apply the bit thread formulation of holographic entanglement entropy to reduced states describing only the geometry contained within an entanglement wedge. We argue that a certain optimized bit thread configuration, which we construct, gives a purification of the reduced state to a full holographic state obeying a precise set of conditional mutual information relations. When this purification exists, we establish, under certain assumptions, the conjectured E P = E W relation equating the entanglement of purification with the area of the minimal cross section partitioning the bulk entanglement wedge. Along the way, we comment on minimal purifications of holographic states, geometric purifications, and black hole geometries.

97 MATHEMATICS AND COMPUTING↗

Effects of Coatings on the High-Cycle Fatigue Life of Threaded Steel Samples

In this work, high-cycle fatigue is studied for threaded cylindrical high-strength steel samples coated using three different industrial processes: black oxidation, normal-temperature galvanization and high-temperature galvanization. The fatigue performance in air is compared with that of uncoated samples. Microstructural characterization revealed the abundant presence of small cracks in the zinc coating partially penetrating into the steel. This is consistent with the observation of multiple crack initiation sites along the thread in the galvanized samples, which led to crescent type fracture surfaces governed by circumferential growth. In contrast, the black oxidized and uncoated samples exhibited a semicircular segment type fracture surface governed by single-sided growth with a significantly longer fatigue life. Numerical fatigue life prediction based on an extended Paris-law formulation has been conducted on two different fracture cases: 2D axisymmetric multisided crack growth and 3D single-sided crack growth. The results of this upper-bound and lower-bound approach are in good agreement with experimental data and can potentially be used to predict the lifetime of bolted components.

36 MATERIALS SCIENCE↗

Threading–Unthreading Transition of Linear-Ring Polymer Blends in Extensional Flow

Adding small amounts of ring polymers to a matrix of their linear counterparts is known to increase the zero-shear-rate viscosity because of linear-ring threading. Uniaxial extensional rheology measurements show that, unlike its pure linear and ring constituents, the blend exhibits an overshoot in the stress growth coefficient. By combining these measurements with ex-situ small-angle neutron scattering and nonequilibrium molecular dynamics simulations, this overshoot is shown here to be driven by a transient threading–unthreading transition of rings embedded within the linear entanglement network. Prior to unthreading, embedded rings deform affinely with the linear entanglement network and produce a measurably stronger elongation of the linear chains in the blend compared to the pure linear melt. Thus, rings uniquely alter the mechanisms of transient elongation in linear polymers.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Self-assembly of aramid amphiphiles into ultra-stable nanoribbons and aligned nanoribbon threads

Small molecule self-assembly is an established route for producing high surface area nanostructures with readily customizable chemistries and precise molecular organization. However, these structures are fragile, exhibiting molecular exchange, migration, and rearrangement - among other dynamic instabilities - and ultimately disassociate upon drying. Here we show a small molecule platform, the aramid amphiphile (AA), that overcomes these dynamic instabilities by incorporating a Kevlar-inspired domain into the molecular structure. Strong, anisotropic interactions between AAs suppress molecular exchange and elicit spontaneous self-assembly in water to form nanoribbons with lengths of up to 20 microns. The nanoribbons have a Young’s modulus of 1.7 GPa and tensile strength of 1.9 GPa. Here, we exploit this stability to extend small molecule self-assembly to hierarchically ordered macroscopic materials outside of solvated environments. Through an aqueous shear alignment process, we organize AA nanoribbons into arbitrarily long flexible threads that support 200 times their weight when dried. Tensile tests of the threads provide a benchmark for Young’s moduli (between ~400 and 600 MPa) and extensibilities (between ~0.6 and 1.1%) that depend on the counterion chemistry in solution. This bottom-up approach to macroscopic materials could benefit solid-state applications, historically inaccessible by self-assembled nanomaterials.

77 NANOSCIENCE AND NANOTECHNOLOGY↗

Reduction of Crosshatch Roughness and Threading Dislocation Density in Metamorphic GaInP Buffers and GaInAs Solar Cells

Surface crosshatch roughness typically develops during the growth of lattice-mismatched compositionally graded buffers and can limit misfit dislocation glide. In this study, the crosshatch roughness during growth of a compressive GaInP/GaAs graded buffer is reduced by increasing the phosphine partial pressure throughout the metamorphic growth. Changes in the average misfit dislocation length are qualitatively determined by characterizing the threading defect density and residual strain. The decrease of crosshatch roughness leads to an increase in the average misfit dislocation glide length, indicating that the surface roughness is limiting dislocation glide. Growth rate is also analyzed as a method to reduce surface crosshatch roughness and increase glide length, but has a more complicated relationship with glide kinetics. Using knowledge gained from these experiments, high quality inverted GaInAs 1 eV solar cells are grown on a GaInP compositionally graded buffer with reduced roughness and threading dislocation density. The open circuit voltage is only 0.38 V lower than the bandgap potential at a short circuit current density of 15 mA/cm{sup 2}, suggesting that there is very little loss due to the lattice mismatch.

14 SOLAR ENERGY↗