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 775 records · Page 43
All sourcesNASA NTRSDOE OSTI
DOE OSTI2019

Accelerating DCA++ (Dynamical Cluster Approximation) Scientific Application on the Summit supercomputer

Optimizing scientific applications on today’s accelerator-based high performance computing systems can be challenging, especially when multiple GPUs and CPUs with heterogeneous memories and persistent non-volatile memories are present. An example is Summit, an accelerator-based system at the Oak Ridge Leadership Computing Facility (OLCF) that is rated as the world’s fastest supercomputer to-date. New strategies are thus needed to expose the parallelism in legacy applications, while being amenable to efficient mapping to the underlying architecture.In this paper we discuss our experiences and strategies to port a scientific application, DCA++, to Summit. DCA++ is a highperformance research application that solves quantum manybody problems with a cutting edge quantum cluster algorithm, the dynamical cluster approximation.Our strategies aim to synergize the strengths of the different programming models in the code. These include: (a) streamlining the interactions between the CPU threads and the GPUs, (b) implementing computing kernels on the GPUs and decreasing CPU-GPU memory transfers, (c) allowing asynchronous GPU communications, and (d) increasing compute intensity by combining linear algebraic operations.Full-scale production runs using all 4600 Summit nodes attained a peak performance of 73.5 PFLOPS with a mixed precision implementation.We observed a perfect strong and weak scaling for the quantum Monte Carlo solver in DCA++, while encountering about 2× input/output (I/O) and MPI communication overhead on the time-to-solution for the full machine run. Our hardware agnostic optimizations are designed to alleviate the communication and I/O challenges observed, while improving the compute intensity and obtaining optimal performance on a complex, hybrid architecture like Summit.

Balduzzi, Giovanni↗
DOE OSTI2018

SharP Data Constructs: Data Constructs to Enable Data-Centric Computing

Extreme-scale applications (i.e., Big-Compute) are becoming increasingly data-intensive, i.e., producing and consuming increasingly large amounts of data. The HPC systems traditionally used for these applications are now used for Big-Data applications such as data analytics, social network analysis, machine learning, and genomics. As a consequence of these trends, the system architecture should be flexible and data-centric. This can already be witnessed in the pre-exascale systems with TBs of on-node hierarchical and heterogeneous memories, PBs of system memory, low-latency, high-throughput networks, and many threaded cores. As such, the pre-exascale systems suit the needs of both Big-Compute and Big-Data applications. Though the system architecture is flexible enough to support both Big-Compute and Big-Data, we argue there is a software gap. Particularly, we need data-centric abstractions to leverage the full potential of the system, i.e., there is a need for native support for data resilience, the ability to express data locality and affinity, mechanisms to reduce data movement, the ability to share data, and abstractions to express User's data usage and data access patterns. In this paper, we (i) show the need for taking a holistic approach towards data-centric abstractions, (ii) show how these approaches were realized in the SHARed data-structure centric Programming abstraction (SharP) library, a data-structure centric programming abstraction, and (iii) apply these approaches to a variety of applications that demonstrate its usefulness. Particularly, we apply these approaches to QMCPack and the Graph500 benchmark and demonstrate the advantages of this approach on extreme-scale systems.

Aderholdt, William Ferrol↗
DOE OSTI2023

AMG Preconditioners based on parallel hybrid coarsening and multi-objective graph matching

We describe preliminary results from a multi-objective graph matching algorithm, in the coarsening step of an aggregation-based Algebraic MultiGrid (AMG) preconditioner, for solving large and sparse linear systems of equations on high-end parallel computers. We have two objectives. First, we wish to improve the convergence behavior of the AMG method when applied to highly anisotropic problems. Second, we wish to extend the parallel package \texttt{PSCToolkit} to exploit multi-threaded parallelism at the node level on multi-core processors. Our matching proposal balances the need to simultaneously compute high weights and large cardinalities by a new formulation of the weighted matching problem combining both these objectives using a parameter $\lambda$. We compute the matching by a parallel $2/3-\varepsilon$-approximation algorithm for maximum weight matchings. Results with the new matching algorithm show that for a suitable choice of the parameter $\lambda$ we compute effective preconditioners in the presence of anisotropy, i.e., smaller solve times, setup times, iterations counts, and operator complexity.

D'Ambra, Pasqua↗
DOE OSTI2021

Towards a III-V Solar Cell with a Metamorphic Graded Buffer Directly Grown on V-Groove Si Substrates

V-groove Si substrates offer a promising route for the direct growth of III-V solar cells on low-cost, photovoltaic-grade Si. In this work, we develop a method for coalescing GaP nucleated on V-groove substrates into thin films suitable for the growth of III-V solar cells. We demonstrate the ability to suppress nucleation-related defects by modifying the V-groove substrate to cover the (0 0 1)-oriented Si V-groove tops with SiN x . A threading dislocation density (TDD) of 5 × 10 7 cm -2 was measured with electron channeling contrast imaging. Continuing work will focus on further reducing the TDD and the growth of a GaAs solar cell on a GaP on V-groove Si template.

41 EE - Solar Energy Technologies Office (EE-4S)↗
DOE OSTI2024

1.1 eV GaInAs Cell Development for Dual-Use Solar and 1070 nm Laser Power Converters

Dual-use power converter cells can receive solar and laser power simultaneously to generate current, an application relevant to space and terrestrial industries. This study investigates two concepts of solar cells optimized for dual-use 1070 nm laser and solar power conversion, a single-junction and triple-junction cell. Because it is bandgap-tuned for a 1070 nm laser, the 1.1 eV junction is incorporated into each of the two designs, making its development key to the success of both concepts. Efficiency data for a one junction GaInAs cell demonstrates a laser conversion efficiency of 38% at 1070nm wavelength without an anti-reflection coating. However, the single-junction device requires optimization to reduce short wavelength absorption of the broad solar spectrum. In both devices, the graded buffer layers in the GaInAs cell affects the cell's performance by reducing threading dislocations in the active junction. However, the buffer in the three-junction device also acts as a lateral transport layer and so affects the fill factor depending on its sheet resistance. By varying the buffer thickness, we demonstrate a direct relationship between buffer thickness and sheet resistance reduction, while considering implications to open-circuit voltages. We also performed resistance modeling to determine the optimal grid spacing and thickness of the grid fingers to minimize losses due to sheet resistance and grid shading.

junctions↗
DOE OSTI2019

CosmoFlow: Using Deep Learning to Learn the Universe at Scale

Deep learning is a promising tool to determine the physical model that describes our universe. To handle the considerable computational cost of this problem, we present CosmoFlow: a highly scalable deep learning application built on top of the TensorFlow framework. CosmoFlow uses efficient implementations of 3D convolution and pooling primitives, together with improvements in threading for many element-wise operations, to improve training performance on Intel ® Xeon Phi™ processors. We also utilize the Cray PE Machine Learning Plugin for efficient scaling to multiple nodes. We demonstrate fully synchronous data-parallel training on 8192 nodes of Cori with 77% parallel efficiency, achieving 3.5 Pflop/s sustained performance. To our knowledge, this is the first large-scale science application of the TensorFlow framework at supercomputer scale with fully-synchronous training. Here, these enhancements enable us to process large 3D dark matter distribution and predict the cosmological parameters Ω M , σ 8 and n s with unprecedented accuracy.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗
DOE OSTI2018

Symmetric Indefinite Linear Solver Using OpenMP Task on Multicore Architectures

Recently, the Open Multi-Processing (OpenMP) standard has incorporated task-based programming, where a function call with input and output data is treated as a task. At run time, OpenMP's superscalar scheduler tracks the data dependencies among the tasks and executes the tasks as their dependencies are resolved. On a shared-memory architecture with multiple cores, the independent tasks are executed on different cores in parallel, thereby enabling parallel execution of a seemingly sequential code. With the emergence of many-core architectures, this type of programming paradigm is gaining attention-not only because of its simplicity, but also because it breaks the artificial synchronization points of the program and improves its thread-level parallelization. In this paper, we use these new OpenMP features to develop a portable high-performance implementation of a dense symmetric indefinite linear solver. Obtaining high performance from this kind of solver is a challenge because the symmetric pivoting, which is required to maintain numerical stability, leads to data dependencies that prevent us from using some common performance-improving techniques. To fully utilize a large number of cores through tasking, while conforming to the OpenMP standard, we describe several techniques. Our performance results on current many-core architectures-including Intel's Broadwell, Intel's Knights Landing, IBM's Power8, and Arm's ARMv8-demonstrate the portable and superior performance of our implementation compared with the Linear Algebra PACKage (LAPACK). The resulting solver is now available as a part of the PLASMA software package.

Computer Science↗
DOE OSTI2022

GSoFa: Scalable Sparse Symbolic LU Factorization on GPUs

Decomposing a matrix $\mathbf {A}$ into a lower matrix $\mathbf {L}$ and an upper matrix $\mathbf {U}$, which is also known as LU decomposition, is an essential operation in numerical linear algebra. For a sparse matrix, LU decomposition often introduces more nonzero entries in the $\mathbf {L}$ and $\mathbf {U}$ factors than in the original matrix. A symbolic factorization step is needed to identify the nonzero structures of $\mathbf {L}$ and $\mathbf {U}$ matrices. Attracted by the enormous potentials of the Graphics Processing Units (GPUs), an array of efforts have surged to deploy various LU factorization steps except for the symbolic factorization, to the best of our knowledge, on GPUs. This article introduces gSoFa, the first GPU-based symbolic factorization design with the following three optimizations to enable scalable LU symbolic factorization for nonsymmetric pattern sparse matrices on GPUs. First, here we introduce a novel fine-grained parallel symbolic factorization algorithm that is well suited for the Single Instruction Multiple Thread (SIMT) architecture of GPUs. Second, we tailor supernode detection into a SIMT friendly process and strive to balance the workload, minimize the communication and saturate the GPU computing resources during supernode detection. Third, we introduce a three-pronged optimization to reduce the excessive space consumption problem faced by multi-source concurrent symbolic factorization. Taken together, gSoFa achieves up to 31× speedup from 1 to 44 Summit nodes (6 to 264 GPUs) and outperforms the state-of-the-art CPU project, on average, by 5×. Notably, gSoFa also achieves up to 47 percent of the peak memory throughput of a V100 GPU in the Summit Supercomputer.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2020

STEPS: A Portable Numerical Simulation Toolkit for Electrical Power System Dynamic Studies

Numerical simulation is the key technique for large scale power system analysis. Redistribution of global renewable power via international interconnections requires new simulation tools to study the interconnected systems with different nominal frequencies as a whole. In this paper we introduce an open source simulation toolkit for electrical power systems (STEPS) which is hosted at Github. Its kernel is coded in C++ with major functions of power flow and electro-mechanical dynamic simulation. Flexible options are provided and configurable to improve power flow solution and dynamic simulation. Common devices and models are supported in STEPS for AC/DC hybrid system studies. Studies of interconnected systems with different nominal frequencies is supported in STEPS for research of international interconnection. Application program interfaces are provided and wrapped with Python to enable high-level interfaces for general applications. STEPS is thread safe and parallel computation is supported in both kernel and script levels to accelerate simulation. It is portable and works on Windows and GNU/Linux platforms. Cases from small to large scale systems are thoroughly tested to validate the toolkit with commercial packages as benchmarks.

42 ENGINEERING↗
DOE OSTI2022

A structural description of the evolution of stakeholders and risk communication in the Department of Energy's defense nuclear facilities: Historical perspective, major stakeholders, and external events

Regulators and policymakers are routinely challenged with explaining complex concepts concerning risk. Part of the challenge is helping external and internal stakeholders to understand the context behind risk-related information and decisions. Here, this paper will describe the historical evolution of the safety and regulatory framework for an important category in the nuclear industry—defense nuclear facilities owned and operated by the US Department of Energy. In parallel with describing this evolution, three major events which occurred external to the complex of defense nuclear facilities will be summarized, and their impact on the maturation of the Department's safety and regulatory framework will be discussed. Finally, integrated with these two threads of discussion will be a chronicle of the changing set of involved organizations and the expanding set of external stakeholders involved in risk decisions—and therefore, the risk communications ecosystem surrounding defense nuclear facilities. It will be noted that this system was once describable as a classic “iron triangle,” but now has progressed to a complex network of federal and state organizations, numerous congressional committees, and expanding sets of external stakeholders. It is hoped that a comprehensive discussion of the context of risk assessment in the defense nuclear facilities complex—addressing historical insights, organizational evolution, and the maturing structure of regulation—will provide enhanced opportunities for building trust and understanding in this complex environment.

98 NUCLEAR DISARMAMENT, SAFEGUARDS, AND PHYSICAL P↗
DOE OSTI2025

1.1 eV GaInAs Cell Development for Dual-Use Solar and 1070 nm Laser Power Converters

Dual-use photovoltaic cells can receive solar and laser power simultaneously to generate current, an application relevant to space and terrestrial industries. This study investigates two concepts of solar cells optimized for dual-use 1070 nm laser and solar power conversion, a single-junction and triple-junction cell. The cell designs are based upon the 3- junction inverted metamorphic solar cell, previously shown highly efficient for solar conversion. Because it is closely bandgap-tuned for a 1070 nm laser, the 1 eV bottom junction is incorporated into each of the two designs, making its development key to the success of both concepts. However, each design requires some modification for efficient dual use. The single-junction device requires optimization to reduce short wavelength absorption of the broad solar spectrum. In both devices, the graded buffer layers in the GaInAs cell affect the cell's performance by reducing threading dislocations in the active junction. However, the buffer in the three-junction device also acts as a lateral transport layer and so affects the fill factor depending on its sheet resistance. By varying the buffer thickness, we demonstrate a direct relationship between buffer thickness and sheet resistance reduction, while considering implications to open-circuit voltages. We also performed resistance modeling to determine the optimal grid spacing and thickness of the grid fingers to minimize losses due to sheet resistance and grid shading. Efficiency data for a one junction GaInAs cell demonstrates a laser conversion efficiency of 38% at 1070 nm wavelength without an anti-reflection coating.

dual-use↗
DOE OSTI2018

Atomic structures of low-complexity protein segments reveal kinked β sheets that assemble networks

Subcellular membraneless assemblies are a reinvigorated area of study in biology, with spirited scientific discussions on the forces between the low-complexity protein domains within these assemblies. To illuminate these forces, we determined the atomic structures of five segments from protein low-complexity domains associated with membraneless assemblies. Their common structural feature is the stacking of segments into kinked β sheets that pair into protofilaments. Unlike steric zippers of amyloid fibrils, the kinked sheets interact weakly through polar atoms and aromatic side chains. By computationally threading the human proteome on our kinked structures, we identified hundreds of low-complexity segments potentially capable of forming such interactions. Finally, these segments are found in proteins as diverse as RNA binders, nuclear pore proteins, and keratins, which are known to form networks and localize to membraneless assemblies.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2022

A DNA molecular printer capable of programmable positioning and patterning in two dimensions

Nanoscale manipulation and patterning usually require costly and sensitive top-down techniques such as those used in scanning probe microscopies or in semiconductor lithography. DNA nanotechnology enables exploration of bottom-up fabrication and has previously been used to design self-assembling components capable of linear and rotary motion. In this work, we combine three independently controllable DNA origami linear actuators to create a nanoscale robotic printer. The two-axis positioning mechanism comprises a moveable gantry, running on parallel rails, threading a mobile sleeve. We show that the device is capable of reversibly positioning a write head over a canvas through the addition of signaling oligonucleotides. We demonstrate “write” functionality by using the head to catalyze a local DNA strand–exchange reaction, selectively modifying pixels on a canvas. Furthermore, this work demonstrates the power of DNA nanotechnology for creating nanoscale robotic components and could find application in surface manufacturing, biophysical studies, and templated chemistry.

77 NANOSCIENCE AND NANOTECHNOLOGY↗
DOE OSTI2017

Peculiarities of growing InGaAs/GaAs/AlGaAs laser structures by MOCVD on Ge/Si substrates

The growth of InGaAs/GaAs/AlGaAs laser structures by MOCVD at low pressures on Si(001) substrates with a Ge epitaxial metamorphic buffer layer of various thicknesses is studied. The results of the influence of the growth temperature and incorporation of additional AlAs layers at the boundary with the Ge/Si(001) substrate on the crystalline and optical quality of the formed III–V structures are presented. It is shown that the incorporation of the AlAs/GaAs/AlAs lattice at the initial growth stages of III–V heterostructures on Ge buffer layers grown on exact Si(001) substrates makes it possible to considerably decrease the density of threading defects and, consequently, form effectively emitting laser structures. The possibility of growing strained InGaAs quantum wells on Si(001) substrates allowing stimulated radiation in a wavelength region longer than 1100 nm is shown.

36 MATERIALS SCIENCE↗
DOE OSTI2018

Effect of Epitaxial-Structure Design and Growth Parameters on the Characteristics of Metamorphic Lasers of the 1.46-μm Optical Range Based on Quantum Dots Grown on GaAs Substrates

The characteristics of lasers of the 1.44–1.46-μm optical range grown on GaAs substrates using a metamorphic buffer are studied. The active region of the laser contains 10 rows of InAs/In{sub 0.4}Ga{sub 0.6}As/In{sub 0.2}Ga{sub 0.8}As quantum dots. It is shown that the use of special selective high-temperature annealing along with the application of short-period In{sub 0.2}Ga{sub 0.8}As/In{sub 0.2}Al{sub 0.3}Ga{sub 0.5}As short-period superlattices makes it possible to substantially decrease the density of threading dislocations in the active region. A threshold current density of 1300 A cm{sup –2}, external differential quantum efficiency of 38%, and maximal pulsed-mode output power of 13 W are attained for a laser with a broad area 3 mm in length.

36 MATERIALS SCIENCE↗
DOE OSTI2018

Reciprocal-Space Maps of X-Ray Diffraction Intensity Distribution and Their Relation to the Dislocation Structure of Epitaxial Layers

X-ray diffraction (XRD) in asymmetric Bragg geometry was measured and XRD intensity distribution maps in the reciprocal space were constructed for GaN epitaxial layers with various degrees of structural perfection grown on c-sapphire substrates. It is established that equal-intensity lines (isolines) related to a regular system of perpendicular rectilinear threading dislocations are extended in a direction parallel to the surface. For a more chaotic distribution of dislocations with a large fraction of horizontal fragments, these isolines are rotated toward a direction perpendicular to the reciprocal lattice vector, although they still not attain a limiting position characteristic of the ideal mosaic crystal.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2018

Asymmetry of the Defect Structure of Semipolar GaN Grown on Si(001)

The defect structure of a thick (~15 μm) semipolar gallium nitride (GaN) layer grown by hydride–chloride vapor phase epitaxy on a Si(001) substrate with buffer layers has been studied by transmission electron microscopy. The asymmetry of the defect structure of GaN epilayer has been revealed and analyzed. The influence of this asymmetry on the rate of decrease in the density of threading dislocations in the growing epitaxial layer is discussed.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗
DOE OSTI2018

Highly scalable distributed-memory sparse triangular solution algorithms.

This paper presents a highly efficient distributed-memory parallel sparse triangular solver. The triangular solution phase is often performed following factorization phase in the sparse linear solvers and has become increasingly computa­tionally expensive for direct solvers with many right hand sides (RHSs) or preconditioned iterative solvers. However, the low arithmetic intensity and sequential nature of the triangular solve algorithm pose performance challenges for its large-scale distributed-memory parallelization. In this work, we propose several strategies to enhance scalability of an algorithm with 2D block cyclic process layout. First, an asynchronous binary-tree-based communication scheme implemented via non-blocking MPI functions is leveraged to broadcast partial solutions and reduce partial updates among a subset of processes for each block column and row of the triangular matrix, respectively. This scheme reduces message latency, improves communication load balance and significantly accelerates asynchronous execution of the triangular solve. In addition, efficient BLAS operations and threading implementations are exploited to accelerate local computations and further reduce process idle time. The proposed strategies are implemented in SuperLU_DIST and numerical experiments show up to 4.4x improvement with one right-hand side and up to 6.1x improvement with 50 right- hand sides on 4096 processes, compared to the current release. This is the first time that sparse triangular solution is demonstrated strong scaling on more than 4000 cores.

Liu, Yang↗