Search NASA⌕ Search

SEARCH · Search NASA

Results for “Parallelization efficiency”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 1,747 records · Page 97

The Reverse Cuthill-McKee Algorithm in Distributed-Memory

Ordering vertices of a graph is key to minimize fill-in and data structure size in sparse direct solvers, maximize locality in iterative solvers, and improve performance in graph algorithms. Except for naturally parallelizable ordering methods such as nested dissection, many important ordering methods have not been efficiently mapped to distributed-memory architectures. In this paper, we present the first-ever distributed-memory implementation of the reverse Cuthill-McKee (RCM) algorithm for reducing the profile of a sparse matrix. Our parallelization uses a two-dimensional sparse matrix decomposition. We achieve high performance by decomposing the problem into a small number of primitives and utilizing optimized implementations of these primitives. Our implementation shows strong scaling up to 1024 cores for smaller matrices and up to 4096 cores for larger matrices.

97 MATHEMATICS AND COMPUTING↗

Bayesian Spectral Moment Estimation and Uncertainty Quantification

We present a Bayesian spectral fitting method developed for spectroscopic data analysis, particularly (but not solely) in the context of fusion energy research. The presented techniques are particularly valuable to estimating moments and corresponding uncertainties whenever the spectra result from line-integrated measurements in nonuniform plasmas, for which the approximation of atomic line shapes being ideal Gaussians gives poor estimates. We decompose multiple, potentially overlapping spectral lines into a sum of Gauss-Hermite polynomials, whose properties allow efficient truncation and uncertainty quantification, often with only three free parameters per atomic emission line. Tests with both synthetic and experimental data demonstrate the effectiveness and robustness where more standard nonlinear fitting routines may experience difficulties. A parallelized version of our implementation is publicly released under an open source license.

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

A-SST Initial Specification

The U.S. Army Research Office (ARO), in partnership with IARPA, are investigating innovative, efficient, and scalable computer architectures that are capable of executing next-generation large scale data-analytic applications. These applications are increasingly sparse, unstructured, non-local, and heterogeneous. Under the Advanced Graphic Intelligence Logical computing Environment (AGILE) program, Performer teams will be asked to design computer architectures to meet the future needs of the DoD and the Intelligence Community (IC). This design effort will require flexible, scalable, and detailed simulation to assess the performance, efficiency, and validity of their designs. To support AGILE, Sandia National Labs will be providing the AGILE-enhanced Structural Simulation Toolkit (A-SST). This toolkit is a computer architecture simulation framework designed to support fast, parallel, and multi-scale simulation of novel architectures. This document describes the A-SST framework, some of its library of simulation models, and how it may be used by AGILE Performers.

97 MATHEMATICS AND COMPUTING↗

Computational design and analysis of modular cells for large libraries of exchangeable product synthesis modules

Microbial metabolism can be harnessed to produce a large library of useful chemicals from renewable resources such as plant biomass. However, it is laborious and expensive to create microbial biocatalysts to produce each new product. To tackle this challenge, we have recently developed modular cell (ModCell) design principles that enable rapid generation of production strains by assembling a modular (chassis) cell with exchangeable production modules to achieve overproduction of target molecules. Previous computational ModCell design methods are limited to analyze small libraries of around 20 products. In this study, we developed a new computational method, named ModCell-HPC, that can design modular cells for large libraries with hundreds of products with a highly-parallel and multi-objective evolutionary algorithm and enable us to elucidate modular design properties. We demonstrated ModCell-HPC to design Escherichia coli modular cells towards a library of 161 endogenous production modules. From these simulations, we identified E. coli modular cells with few genetic manipulations that can produce dozens of molecules in a growth-coupled manner with different types of fermentable sugars. These designs revealed key genetic manipulations at the chassis and module levels to accomplish versatile modular cells, involving not only in the removal of major by-products but also modification of branch points in the central metabolism. We further found that the effect of various sugar degradation on redox metabolism results in lower compatibility between a modular cell and production modules for growth on pentoses than hexoses. To better characterize the degree of compatibility, we developed a method to calculate the minimal set cover, identifying that only three modular cells are all needed to couple with up 85 compatible production modules. By determining the unknown compatibility contribution metric, we further elucidated the design features that allow an existing modular cell to be re-purposed towards production of new molecules. Altogether, ModCell-HPC is a useful tool for understanding modularity of biological systems and guiding more efficient and generalizable design of modular cells that help reduce research and development cost in biocatalysis.

59 BASIC BIOLOGICAL SCIENCES↗

Development of human machine interface (HMI) for digital control rooms in nuclear power plants - 119

The analog control systems of traditional Nuclear Power Plants (NPPs) are being replaced by digital systems due to the fact that analog Instrumentation and Control (I and C) systems are becoming obsolete and such technology is no longer supported by manufacturers. The purpose of this study is to jointly evaluate the performance of an operator with a Human Machine Interface (HMI) in a NPP digital control room ensuring the efficiency and the safety of the NPP. The validity and the accuracy of the operators' actions are the primary goals that have to be preserved through the emergence of the HMI based in digital systems. The transition from analog to digital control systems leads to a more advanced support of the NPP; Video Display Units (VDUs) are used for monitoring the state of the NPP while alarm systems help for directing attention. The operators, in order to monitor information from the different VDUs, have to switch attention between different screens (alarms systems and control systems). Therefore, a well-established HMI is a significant challenge to the overall operation of the NPP, that actually helps the operators to learn from the available information in detecting possible anomalies and making the right decisions, to ensure safety, and in parallel to diminish the human error. (authors)

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

Automatic Halo Management for the Uintah GPU-Heterogeneous Asynchronous Many-Task Runtime

The Uintah computational framework is used for the parallel solution of partial differential equations on adaptive mesh refinement grids using modern supercomputers. Uintah is structured with an application layer and a separate runtime system. Uintah is based on a distributed directed acyclic graph of computational tasks, with a task scheduler that efficiently schedules and executes these tasks on both CPU cores and on-node accelerators. The runtime system identifies task dependencies, creates a task graph prior to the execution of these tasks, automatically generates MPI message tags, and automatically performs halo transfers for simulation variables. Automating halo transfers in a heterogeneous environment poses significant challenges when tasks compute within a few milliseconds, as runtime overhead affects wall time execution, or when simulation variables require large halos spanning most or all of the computational domain, as task dependencies become expensive to process. These challenges are magnified at production scale when application developers require each compute node perform thousands of different halo transfers among thousands simulation variables. The principal contribution of this work is to (1) identify and address inefficiencies that arise when mapping tasks onto the GPU in the presence of automated halo transfers, (2) implement new schemes to reduce runtime system overhead, (3) minimize application developer involvement with the runtime, and (4) show overhead reduction results from these improvements.

Concurrency↗

A high-speed one-dimensional detector for time-resolved small -angle X-ray scattering : design and characterization.

A high-speed one-dimensional detector for time-resolved small-angle x-ray scattering has been designed and built for experiments at the Advanced Photon Source of Argonne National Laboratory. This detector is made from a 500-?m thick by 150-mm diameter ultra-high-purity n-type silicon wafer. The electrodes, which are a series of concentric rings that are deposited in the wafer, integrate the scattered x-rays over the azimuthal angle and, thereby, produce a one-dimensional detector. This design yields 128 rings, which allows parallel processing of the signal from each ring. The readout electronics consist of transimpedance front-end amplifiers, one for each ring, followed by active pulse-shaping filters. The amplifier signals are digitized using 12-bit analog-to-digital converters, one per ring, which operate at 20 MHz. The frame rate of the system is 271 kHz. Up to 220 - 1 scattering profiles may be stored on a random access memory chip and transferred to a data file at a rate of 16 x 103 profiles/sec. For X-ray energies between 3.5 and 13.2 keV the efficiency exceeds 80%. The resolving time of the electronics is 300 ns, which is sufficient to isolate electronically a single pulse of scattered x-rays when the synchrotron is operated in a hybrid or asymmetric fill pattern. Therefore, laser-pump/x-ray-probe experiments can be performed without a mechanical shutter. Examples of time-resolved speckle and the kinetics of the formation of sodium chloride particles are presented. This detector is capable of acquiring small-angle x-ray scattering profiles over multiple time scales, which are needed to characterize many chemical, physical, and biological processes. In addition, this detector may be tested and calibrated before experimental runs, without access to an intense beam of x-rays, with alpha particles from a radioactive source such as 241Am.

43 PARTICLE ACCELERATORS↗

TORCH Computational Reference Kernels - A Testbed for Computer Science Research

For decades, computer scientists have sought guidance on how to evolve architectures, languages, and programming models in order to improve application performance, efficiency, and productivity. Unfortunately, without overarching advice about future directions in these areas, individual guidance is inferred from the existing software/hardware ecosystem, and each discipline often conducts their research independently assuming all other technologies remain fixed. In today's rapidly evolving world of on-chip parallelism, isolated and iterative improvements to performance may miss superior solutions in the same way gradient descent optimization techniques may get stuck in local minima. To combat this, we present TORCH: A Testbed for Optimization ResearCH. These computational reference kernels define the core problems of interest in scientific computing without mandating a specific language, algorithm, programming model, or implementation. To compliment the kernel (problem) definitions, we provide a set of algorithmically-expressed verification tests that can be used to verify a hardware/software co-designed solution produces an acceptable answer. Finally, to provide some illumination as to how researchers have implemented solutions to these problems in the past, we provide a set of reference implementations in C and MATLAB.

97 MATHEMATICS AND COMPUTING↗

A scalable framework for efficient coupling of thermal and microstructural simulations in additive manufacturing

Predicting microstructure evolution in metal additive manufacturing (AM) is important for process optimization, but spatiotemporal scale disparities between thermal transport and microstructure evolution create significant challenges for efficient data transfer between simulation codes. To address this, we present Stork, a scalable framework for coupling thermal and microstructural simulations. Stork uses a sparse data representation to identify and store active solidification sub-volumes, enabling highly parallel quad-linear interpolation from coarse thermal grids to fine microstructure grids without large intermediate storage. We demonstrate the framework by coupling the semi-analytic heat transfer code 3DThesis with the time-parallel cellular automata code Toucan. This approach achieves over two orders of magnitude reduction in data generation time and file size compared to prior workflows. Numerical studies show that quad-linear interpolation preserves grain morphology and crystallographic texture in laser powder bed fusion (LPBF) simulations for coarsening ratios up to 16. Overall, Stork provides a scalable pathway for high-throughput, component-scale AM simulations on modern high-performance computing systems.

36 MATERIALS SCIENCE↗

Scaling Deep Learning workloads: NVIDIA DGX-1/Pascal and Intel Knights Landing

Deep Learning (DL) algorithms have become ubiquitous in data analytics. As a result, major computing vendors—including NVIDIA, Intel, AMD, and IBM—have architectural road maps influenced by DL workloads. Furthermore, several vendors have recently advertised new computing products as accelerating large DL workloads. Unfortunately, it is difficult for data scientists to quantify the potential of these different products. Here, this article provides a performance and power analysis of important DL workloads on two major parallel architectures: NVIDIA DGX-1 (eight Pascal P100 GPUs interconnected with NVLink) and Intel Knights Landing (KNL) CPUs interconnected with Intel Omni-Path or Cray Aries. Our evaluation consists of a cross section of convolutional neural net workloads: CifarNet, AlexNet, GoogLeNet, and ResNet50 topologies using the Cifar10 and ImageNet datasets. The workloads are vendor-optimized for each architecture. We use sequentially equivalent implementations to maintain iso-accuracy between parallel and sequential DL models. Our analysis indicates that although GPUs provide the highest overall performance, the gap can close for some convolutional networks; and the KNL can be competitive in performance/watt. We find that NVLink facilitates scaling efficiency on GPUs. However, its importance is heavily dependent on neural network architecture. Furthermore, for weak-scaling—sometimes encouraged by restricted GPU memory—NVLink is less important.

97 MATHEMATICS AND COMPUTING↗

Current and Future Performance of the CMS Simulation

The CMS full simulation using Geant4 has delivered billions of simulated events for analysis during Runs 1 and 2 of the LHC. However, the HL-LHC dataset will be an order of magnitude larger, with a similar increase in occupancy per event. In addition, the upgraded CMS detector will be considerably more complex, with an extended silicon tracker and a high granularity calorimeter in the endcap region. Increases in conventional computing resources are subject to both technological and budgetary limitations, so novel approaches are needed to improve software efficiency and to take advantage of new architectures and heterogeneous resources. Several projects are in development to address these needs, including the vectorized geometry library Vec-Geom and the GeantV transport engine, which uses track-level parallelization. The current computing performance of the CMS simulation will be presented as a baseline, along with an overview of the various optimizations already available for Geant4. Finally, the progress and outlook for integrating VecGeom and GeantV in the CMS software framework will be discussed.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

2016 High Performance Computing Annual Report

Sandia National Laboratories’ California site is celebrating its 60th anniversary (1956 to 2016), and high performance computing has been a key enabler for its scientists and engineers throughout much of its history. Since its founding, Sandia California has helped pioneer the use of HPC platforms including hosting Sandia’s first Cray-1 supercomputer in the 1970s and supporting development of scalable cluster computing platforms to create a new paradigm for cost-effective supercomputing in the 1990s. Recent decades of investment in creation of scalable application frameworks for scientific computing have also enabled new generations of modeling and simulation codes. These resources have facilitated computational analysis of complex phenomena in diverse applications spanning national defense, energy, and homeland security. Today, Sandia California researchers work with partners in academia, industry, and national labs to evolve the state-of-the-art in HPC, modeling, and data analysis (including foundational capabilities for exascale computing platforms) and apply them in transformational ways. Research efforts include mitigating the effects of silent hardware failures that can jeopardize the results of large-scale computations, developing exascale-capable asynchronous task-parallel programming models and runtime systems, formulating new techniques to better explore and analyze extreme-scale data sets, and increasing our understanding of materials and chemical sciences which has applications spanning nuclear weapons stockpile stewardship to more efficient automobile engines. The following section highlights some of these research and applications projects and further illustrates the breadth of our HPC capabilities.

97 MATHEMATICS AND COMPUTING↗

NEAMS Technical Area Support in MOOSE

The Multiphysics Object-Oriented Simulation Environment (MOOSE) framework is a foundational capability used by the Nuclear Energy Advanced Modeling and Simulation (NEAMS) program to create over 15 different simulation tools for advanced nuclear reactors. Due to this ubiquity, improvements to the framework in support of modeling and simulation goals are critical to the program. These improvements can take many forms, including optimization, improved user experience, streamlined application programming interfaces (APIs), parallelism, and other new capabilities. The work described in this report was conducted in direct support of the simulation tools and has already been deployed. The capabilities outlined in this report include implementing hash table matrix assembly for efficient sparsity pattern construction for contact in BISON, developing re-step testing infrastructure for ensuring the viability of overlapping domain coupling between SAM and Pronghorn, allowing unique preconditioners for single-input multi-system solves, supporting multi-system in MOOSE’s workhorse executioners, and many more smaller feature enhancements and bug fixes.

97 - MATHEMATICS AND COMPUTING↗

Bright Gamma-Ray Flares Powered by Magnetic Reconnection in QED-strength Magnetic Fields

Strong magnetic fields in the magnetospheres of neutron stars (NSs) (especially magnetars) and other astrophysical objects may release their energy in violent, intense episodes of magnetic reconnection. While reconnection has been studied extensively, the extreme field strength near NSs introduces new effects: radiation cooling and electron–positron pair production. Using massively parallel particle-in-cell simulations that self-consistently incorporate these new radiation and quantum-electrodynamic effects, we investigate relativistic magnetic reconnection in the strong-field regime. We show that reconnection in this regime can efficiently convert magnetic energy to X-ray and gamma-ray radiation and thus power bright, high-energy astrophysical flares. Rapid radiative cooling causes strong plasma and magnetic field compression in compact plasmoids. In the most extreme cases, the field can approach the quantum limit, leading to copious pair production.

79 ASTRONOMY AND ASTROPHYSICS↗

RAYLEIGH QUOTIENT ITERATION IN 3D, DETERMINISTIC NEUTRON TRANSPORT

Today’s “grand challenge” neutron transport problems require 3-D meshes with billions of cells, hundreds of energy groups, and accurate quadratures and scattering expansions. Leadership-class computers provide platforms on which high-fidelity fluxes can be calculated. However, appropriate methods are needed that can use these machines effectively. Such methods must be able to to use hundreds of thousands of cores and have good convergence properties. Rayleigh quotient iteration (RQI) is an eigenvalue solver that has been added to the SN code Denovo to address convergence. Rayleigh quotient iteration is an optimal shifted inverse iteration method that should converge in fewer iterations than the more common power method and other shifted inverse iteration methods for many problems of interest. Denovo’s RQI uses a new multigroup Krylov solver for the fixed source solutions inside every iteration that allows parallelization in energy in addition to space and angle. This Krylov solver has been shown to scale successfully to 200,000 cores: for example one test problem scaled from 69,120 cores to 190,080 cores with 98% efficiency. This paper shows that RQI works for some small problems. However, the Krylov method upon which it relies does not always converge because RQI creates ill-conditioned systems. This result leads to the conclusion that preconditioning is needed to allow this method to be applicable to a wider variety of problems.

Slaybaugh, R. N.↗

Rayleigh Quotient Iteration in 3D, Deterministic Neutron Transport

Today's "grand challenge" neutron transport problems require 3-D meshes with billions of cells, hundreds of energy groups, and accurate quadratures and scattering expansions. Leadership-class computers provide platforms on which high-fidelity fluxes can be calculated. However, appropriate methods are needed that can use these machines effectively. Such methods must be able to use hundreds of thousands of cores and have good convergence properties. Rayleigh quotient iteration (RQI) is an eigenvalue solver that has been added to the Sn code Denovo to address convergence. Rayleigh quotient iteration is an optimal shifted inverse iteration method that should converge in fewer iterations than the more common power method and other shifted inverse iteration methods for many problems of interest. Denovo's RQI uses a new multigroup Krylov solver for the fixed source solutions inside every iteration that allows parallelization in energy in addition to space and angle. This Krylov solver has been shown to scale successfully to 200,000 cores: for example one test problem scaled from 69,120 cores to 190,080 cores with 98% efficiency. This paper shows that RQI works for some small problems. However, the Krylov method upon which it relies does not always converge because RQI creates ill-conditioned systems. This result leads to the conclusion that preconditioning is needed to allow this method to be applicable to a wider variety of problems.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

A Screen Space GPGPU Surface LIC Algorithm for Distributed Memory Data Parallel Sort Last Rendering Infrastructures

The surface line integral convolution(LIC) visualization technique produces dense visualization of vector fields on arbitrary surfaces. We present a screen space surface LIC algorithm for use in distributed memory data parallel sort last rendering infrastructures. The motivations for our work are to support analysis of datasets that are too large to fit in the main memory of a single computer and compatibility with prevalent parallel scientific visualization tools such as ParaView and VisIt. By working in screen space using OpenGL we can leverage the computational power of GPUs when they are available and run without them when they are not. We address efficiency and performance issues that arise from the transformation of data from physical to screen space by selecting an alternate screen space domain decomposition. We analyze the algorithm's scaling behavior with and without GPUs on two high performance computing systems using data from turbulent plasma simulations.

Loring, Burlen↗