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 1,207 records · Page 67

Exascale and Beyond: Configuring, Reasoning, Scaling. Report of the 2011 Workshop on Architectures II: Exascale and Beyond, held August 8-10, Albuquerque, N.M.

The push toward exascale computing (1) and beyond is torn between contradictory and possibly incompatible goals. On the one hand, one wants a system that will be available within a short span of time, on a relatively small development budget, and with the least changes possible to the current software stack. On the other hand, one wants a system that is orders of magnitude more efficient in its energy consumption. An exascale system could be as different from current systems as scalable parallel systems were from vector machines at the time of the last major transition in the field, but the amount of application software developed at DOE in the last decades makes it undesirable to take a “clean sheet of paper” approach, if at all possible. Furthermore, one wishes to explore and develop distinct exascale architectures to reduce risk while maintaining some unity so that application codes easily port among them. Finally, the design of an exascale system may involve a large number of vendors, some of which have limited experience in the field of High-Performance Computing (HPC). A proper integration of these multiple contributions requires a proper definition of many new interfaces, as well as consistent design principles for the handling of cross-cutting issues such as resilience, performance, and power management.

97 MATHEMATICS AND COMPUTING↗

MAESTROeX: A Massively Parallel Low Mach Number Astrophysical Solver

@e present MAESTROeX, a massively parallel solver for low Mach number astrophysical flows. The underlying low Mach number equation set allows for efficient, long-time integration for highly subsonic flows compared to compressible approaches. MAESTROeX is suitable for modeling full spherical stars as well as well as planar simulations of dynamics within localized regions of a star, and can robustly handle several orders of magnitude of density and pressure stratification. Previously, we have described the development of the predecessor of MAESTROeX, called MAESTRO, in a series of papers. Here, we present a new, greatly simplified temporal integration scheme that retains the same order of accuracy as our previous approaches. We also explore the use of alternative spatial mapping of the one-dimensional base state onto the full Cartesian grid. The code leverages the new AMReX software framework for block-structured adaptive mesh refinement (AMR) applications, allowing for scalability to large fractions of leadership-class machines. Finally, using our previous studies on the convective phase of single-degenerate progenitor models of SNe Ia as a guide, we characterize the performance of the code and validate the new algorithmic features. Like MAESTRO, MAESTROeX is fully open source.

79 ASTRONOMY AND ASTROPHYSICS↗

Hybrid MPI

Enables parallel applications wrriten to use the Message Passing Interface Standard to communicate across different processors to more efficiently leverage shared memory hardware devices.

Source record↗

Expert and operator perspectives on barriers to energy efficiency in data centers

Abstract It was last estimated in 2016 that data centers (DCs) comprise approximately 2% of total US electricity consumption. However, this estimate is currently being updated to account for the massive increase in computing needs due to streaming, cryptocurrency, and artificial intelligence (AI). To prevent energy consumption that tracks with increasing computing needs, it is imperative we identify energy efficiency strategies and investments beyond the low-hanging fruit solutions. In a two-phased research approach, we ask: What non-technical barriers still impede energy efficiency (EE) practices and investments in the data center sector, and what can be done to overcome these barriers? In particular, we are focused on social and organizational barriers to EE. In Phase I, we performed a literature review and found that technical solutions are abundant in the literature, but fail to address the top-down cultural shifts that need to take place in order to adapt new energy efficiency strategies. In Phase II, reported here, we interviewed 16 data center operators/experts to ground-truth our literature findings. Our interview protocols focus on three aspects of DC decision-making: procurement practices, metrics and monitoring, and perceived barriers to energy efficiency. We find that vendors are the key drivers of procurement decisions, advanced efficiency metrics are facility-specific, and there is convergence in the design of advanced facilities due to the heat density of parallelized infrastructure. Our ultimate goals for our research are to design DC decarbonization policies that target organizational structure, empower individual staff, and foster a supportive external market.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Multithreaded Algorithms for Maximum Matching in Bipartite Graphs

Abstract—We design, implement, and evaluate algorithms for computing a matching of maximum cardinality in a bipartite graph on multi-core and massively multithreaded computers. As computers with larger number of slower cores dominate the commodity processor market, the design of multithreaded algorithms to solve large matching problems becomes a necessity. Recent work on serial algorithms based on searching for augmenting paths for this problem have shown that their performance is sensitive to the order in which the vertices are processed for matching. In a multithreaded environment, imposing a serial order in which vertices are considered for matching would lead to loss of concurrency and performance. But this raises the question: Would parallel matching algorithms on multithreaded machines improve performance over a serial algorithm? We answer this question in the affirmative. We report efficient multithreaded implementations of two key algorithms (Hopcroft- Karp based on breadth-first-search, and Pothen-Fan based on depth-first-search) and their variants, combined with the Karp- Sipser initialization algorithm. We report extensive results and insights using three shared-memory platforms (a 48-core AMD Opteron, a 32-core Intel Nehalem, and a 128-processor Cray XMT) on a representative set of real-world and synthetic graphs. To the best of our knowledge, this is the first extensive study of augmentation-based parallel algorithms for bipartite cardinality matching.

Azad, Md Ariful↗

Solar Field Layout and Aimpoint Strategy Optimization

The existing methods that determine heliostat aiming strategies for concentrating solar power (CSP) central receiver plants typically use heuristics and/or are computationally expensive, and they lack flexibility for different desired flux profiles and receiver geometries. Because of the interaction between layout and aimpoint strategy, considering the former without accounting for the latter may yield solutions with superfluous heliostats that cannot be used efficiently without compromising receiver flux constraints. To that end, we develop a software decision tool that uses innovative optimization methods to both optimize aimpoint strategies and improve candidate layouts for the solar collection field of a CSP central receiver plant. A CSP plant’s effectiveness relies on the optical efficiency of the solar field, which may be limited by losses due to (i) the cosine effect, (ii) atmospheric attenuation, (iii) interference (i.e., shading and blocking) between heliostats, (iv) spillage as a result of heliostat positioning and geometry, and (iv) some heliostats’ inability to direct irradiance to the receiver without damage due to excessive thermal flux. The goal of this work is to obtain optimized aiming strategies and improved solar field layouts that reduce capital cost and increase field optical efficiency and utilization, while meeting the power requirements of a given CSP receiver design. We formulate the aimpoint optimization problem as a mixed-integer linear programming model, which we then decompose into submodels that we solve in parallel. The decomposition subdivides the solar field into sections, and aimpoint strategies for each section are obtained independently of the others. To improve existing layouts, we develop a utilization-weighted efficiency metric that we use to relocate heliostats to sections of the solar field with similar efficiency and higher utilization. Finally, to connect our software to high-fidelity flux models, we develop a Python application programming interface for SolarPILOT, a mature software package that characterizes solar field performance and generates the heliostat layouts and flux maps that serve as input to our models.

14 SOLAR ENERGY↗

A novel method for experimental characterization of large-angle scattered particles in scanned carbon-ion therapy

Purpose: It is essential to consider large-angle scattered particles in dose calculation models for therapeutic carbon-ion beams. However, it is difficult to measure the small dose contribution from large-angle scattered particles. In this paper, the authors present a novel method to derive the parameters describing large-angle scattered particles from the measured results. Methods: The authors developed a new parallel-plate ionization chamber consisting of concentric electrodes. Since the sensitive volume of each channel is increased linearly with this type, it is possible to efficiently and easily detect small contributions from the large-angle scattered particles. The parameters describing the large-angle scattered particles were derived from pencil beam dose distribution in water measured with the new ionization chamber. To evaluate the validity of this method, the correction for the field-size dependence of the doses, “predicted-dose scaling factor,” was calculated with the new parameters. Results: The predicted-dose scaling factor calculated with the new parameters was compared with the existing one. The difference between the new correction factor and the existing one was 1.3%. For target volumes of different sizes, the calculated dose distribution with the new parameters was in good agreement with the measured one. Conclusions: Parameters describing the large-angle scattered particles can be efficiently and rapidly determined using the new ionization chamber. The authors confirmed that the field-size dependence of the doses could be compensated for by the new parameters. This method makes it possible to easily derive the parameters describing the large-angle scattered particles, while maintaining the dose calculation accuracy.

61 RADIATION PROTECTION AND DOSIMETRY↗

An Optimization Model for Plug-In Hybrid Electric Vehicles

The necessity for environmentally conscious vehicle designs in conjunction with increasing concerns regarding U.S. dependency on foreign oil and climate change have induced significant investment towards enhancing the propulsion portfolio with new technologies. More recently, plug-in hybrid electric vehicles (PHEVs) have held great intuitive appeal and have attracted considerable attention. PHEVs have the potential to reduce petroleum consumption and greenhouse gas (GHG) emissions in the commercial transportation sector. They are especially appealing in situations where daily commuting is within a small amount of miles with excessive stop-and-go driving. The research effort outlined in this paper aims to investigate the implications of motor/generator and battery size on fuel economy and GHG emissions in a medium-duty PHEV. An optimization framework is developed and applied to two different parallel powertrain configurations, e.g., pre-transmission and post-transmission, to derive the optimal design with respect to motor/generator and battery size. A comparison between the conventional and PHEV configurations with equivalent size and performance under the same driving conditions is conducted, thus allowing an assessment of the fuel economy and GHG emissions potential improvement. The post-transmission parallel configuration yields higher fuel economy and less GHG emissions compared to pre-transmission configuration partly attributable to the enhanced regenerative braking efficiency.

02 PETROLEUM↗

CP2K: An Electronic Structure and Molecular Dynamics Software Package - Quickstep: Efficient and Accurate Electronic Structure Calculations

CP2K is an open source electronic structure and molecular dynamics software package to perform atomistic simulations of solid-state, liquid, molecular and biological systems. It is especially aimed at massively-parallel and linear-scaling electronic structure methods and state-of-the-art ab-initio molecular dynamics simulations. Excellent performance for electronic structure calculations is achieved using novel algorithms implemented for modern high-performance computing systems. This review revisits the main capabilities of CP2K to perform efficient and accurate electronic structure simulations. The emphasis is put on density functional theory and multiple post-Hartree-Fock methods using the Gaussian and plane wave approach and its augmented all-electron extension. TDK has received funding from the European Research Council (ERC) under the European Union's Horizon 2020 research and innovation programme (grant agreement No. 716142). VRR has been supported by the Swiss National Science Foundation in the form of Ambizione grant No. PZ00P2 174227 and RZK by the Natural Sciences and Engineering Research Council of Canada (NSERC) through Discovery Grants (RGPIN-2016-0505). GKS and CJM are supported by the US Department of Energy, Office of Science, Office of Basic Energy Sciences, Division of Chemical Sciences, Geosciences, and Biosciences. UK based work was funded under the embedded CSE programme of the ARCHER UK National Supercomputing Service (http://www.archer.ac.uk), grants eCSE03-011, eCSE06-6, eCSE08-9, eCSE13-17 and the EPSRC (EP/P022235/1) grant “Surface and Interface Toolkit for the Materials Chemistry Community". Computational resources were provided by the Swiss National Supercomputing Centre (CSCS) and Compute Canada. The generous allocation of computing time on the FPGA-based supercomputer “Noctua" at PC2 is kindly acknowledged.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Modeling UV Radiation Feedback from Massive Stars. I. Implementation of Adaptive Ray-tracing Method and Tests [Modeling UV Radiation Feedback from Massive Stars: I. Implementation of Adaptive Ray Tracing Algorithm]

Here, we present an implementation of an adaptive ray-tracing (ART) module in the Athena hydrodynamics code that accurately and efficiently handles the radiative transfer involving multiple point sources on a three-dimensional Cartesian grid. We adopt a recently proposed parallel algorithm that uses nonblocking, asynchronous MPI communications to accelerate transport of rays across the computational domain. We validate our implementation through several standard test problems, including the propagation of radiation in vacuum and the expansions of various types of H II regions. Additionally, scaling tests show that the cost of a full ray trace per source remains comparable to that of the hydrodynamics update on up to $\sim {10}^{3}$ processors. To demonstrate application of our ART implementation, we perform a simulation of star cluster formation in a marginally bound, turbulent cloud, finding that its star formation efficiency is 12% when both radiation pressure forces and photoionization by UV radiation are treated. We directly compare the radiation forces computed from the ART scheme with those from the M 1 closure relation. Although the ART and M 1 schemes yield similar results on large scales, the latter is unable to resolve the radiation field accurately near individual point sources.

79 ASTRONOMY AND ASTROPHYSICS↗

A New Configuration of Paralleled Modular ANPC Multilevel Converter Controlled by an Improved Modulation Method for 1 MHz, 1 MW EV Charger

In this work, a new configuration of the modular multilevel converter (MLC) based on the parallel connection of three-level active-neutral-point-clamped (3L-ANPC) cells as well as its improved modulation method is proposed for 1 MHz, 1 MW electric vehicle (EV) megacharger. In the proposed paralleled modular ANPC-MLC, only six high-frequency silicon carbide (SiC) power switches operating at 333 kHz are required to generate 1 MHz switching frequency spectrum. Moreover, the operating voltage of all power devices is halved, the magnitude of the first switching frequency harmonic cluster is decreased by the factor of five, and the load current is equally distributed between the 3L-ANPC legs by employing the proposed improved modulation method. Hence, the modularity, efficiency, and power density of the proposed converter are notably increased, whereas the value of passive components and the overall switching loss are remarkably decreased. In addition, an optimized design of the one 3L-ANPC cell of the proposed paralleled modular ANPC-MLC for 1 MHz, 1MW EV megacharger using Ansys SIwave, Icepak, and Q3D finite element method platforms is presented and analyzed in detail. The provided experimental results of the down-scaled setup verify the feasibility and viability of the proposed configuration as well as its improved switching pattern.

42 ENGINEERING↗

LDRD Final Report (08-ERD-037): Important Modes to Drive Protein MD Simulations to the Next Conformational Level

Every action in biology is performed by dynamic proteins that convert between multiple states in order to engage their functions. Often binding to various ligands is essential for the rates of desired transitions to be enhanced. The goal of computational biology is to study these transitions and discover the different states to fully understand the protein's normal and diseased function, design drugs to target/bias specific states, and understand all of the interactions in between. We have developed a new methodology that is capable of calculating the absolute free energy of proteins while taking into account all the interactions with the solvent molecules. The efficiency of the new scheme is an order of magnitude greater than any existing technique. This method is now implemented in the massively parallel popular MD program package NAMD. This now makes it possible to calculate the relative stability of different conformational states of biological macromolecules as well as their binding free energies to various ligands.

59 BASIC BIOLOGICAL SCIENCES↗

C-SAW: a framework for graph sampling and random walk on GPUs

Many applications require to learn, mine, analyze and visualize large-scale graphs. These graphs are often too large to be addressed efficiently using conventional graph processing technologies. Fortunately, recent research efforts find out graph sampling and random walk, which significantly reduce the size of original graphs, can benefit the tasks of learning, mining, analyzing and visualizing large graphs by capturing the desirable graph properties. This paper introduces C-SAW, the first framework that accelerates Sampling and Random Walk framework on GPUs. Particularly, C-SAW makes three contributions: First, our framework provides a generic API which allows users to implement a wide range of sampling and random walk algorithms with ease. Second, offloading this framework on GPU, we introduce warp-centric parallel selection, and two novel optimizations for collision migration. Third, towards supporting graphs that exceed the GPU memory capacity, we introduce efficient data transfer optimizations for out-of-memory and multi-GPU sampling, such as workload-aware scheduling and batched multi-instance sampling. Taken together, our framework constantly outperforms the state of the art projects in addition to the capability of supporting a wide range of sampling and random walk algorithms.

97 MATHEMATICS AND COMPUTING↗

Efficient Unsteady Flow Visualization with High-Order Access Dependencies

We present a novel high-order access dependencies based model for efficient pathline computation in unsteady flow visualization. By taking longer access sequences into account to model more sophisticated data access patterns in particle tracing, our method greatly improves the accuracy and reliability in data access prediction. In our work, high-order access dependencies are calculated by tracing uniformly-seeded pathlines in both forward and backward directions in a preprocessing stage. The effectiveness of our proposed approach is demonstrated through a parallel particle tracing framework with high-order data prefetching. Results show that our method achieves higher data locality and hence improves the efficiency of pathline computation.

Zhang, Jiang↗

The ecology of flows and drift wave turbulence in CSDX: A model

This paper describes the ecology of drift wave turbulence and mean flows in the coupled drift-ion acoustic wave plasma of a CSDX linear device. A 1D reduced model that studies the spatiotemporal evolution of plasma mean density $\bar{n}$, and mean flows $\bar{v}$y and $\bar{v}$z, in addition to fluctuation intensity ε, is presented. Here, ε = 〈 n ~ 2 + ( ∇ ⊥ Φ ~ ) 2 + v ~ z 2 〉 is the conserved energy field. The model uses a mixing length l mix inversely proportional to both axial and azimuthal flow shear. This form of l mix closes the loop on total energy. The model self-consistently describes variations in plasma profiles, including mean flows and turbulent stresses. It investigates the energy exchange between the fluctuation intensity and mean profiles via particle flux &#9001 n ~ v ~ x 〉 and Reynolds stresses 〈 v ~ x v ~ y 〉 and 〈 v ~ x v ~ z 〉 Acoustic coupling breaks parallel symmetry and generates a parallel residual stress Π res xz . The model uses a set of equations to explain the acceleration of $\bar{v}_y$ and $\bar{v}_z$ via Π x y r e s ∝ ∇ n - and Π x y r e s ∝ ∇ n - . Flow dynamics in the parallel direction are related to those in the perpendicular direction through an empirical coupling constant σVT. This constant measures the degree of symmetry breaking in the 〈$k_mk_z$〉 correlator and determines the efficiency of ∇$\bar{n}$ in driving $\bar{v}_z$. The model also establishes a relation between ∇$\bar{v}_y$ and ∇$\bar{v}_z$, via the ratio of the stresses Π res xy and Π res xz . When parallel to perpendicular flow coupling is weak, axial Reynolds power P x z R e = - 〈 v ~ x v ~ z &#9002 ∇ v - z is less than the azimuthal Reynolds power P x y R e = - 〈 v ~ x v ~ y 〉 ∇ v - y ∇$\bar{v}_y$. The model is then reduced to a 2-field predator/prey model where $\bar{v}_z$ is parasitic to the system and fluctuations evolve self-consistently. Finally, turbulent diffusion in CSDX follows the scaling: D CSDX = D B ρ * 0.6 , where D B is the Bohm diffusion coefficient and ρ * is the ion gyroradius normalized to the density gradient | ∇ n - / n - | - 1

70 PLASMA PHYSICS AND FUSION TECHNOLOGY↗

Singular value decomposition utilizing parallel algorithms on graphical processors

One of the current challenges in underwater acoustic array signal processing is the detection of quiet targets in the presence of noise. In order to enable robust detection, one of the key processing steps requires data and replica whitening. This, in turn, involves the eigen-decomposition of the sample spectral matrix, Cx = 1/K xKX(k)XH(k) where X(k) denotes a single frequency snapshot with an element for each element of the array. By employing the singular value decomposition (SVD) method, the eigenvectors and eigenvalues can be determined directly from the data without computing the sample covariance matrix, reducing the computational requirements for a given level of accuracy (van Trees, Optimum Array Processing). (Recall that the SVD of a complex matrix A involves determining V, , and U such that A = U VH where U and V are orthonormal and is a positive, real, diagonal matrix containing the singular values of A. U and V are the eigenvectors of AAH and AHA, respectively, while the singular values are the square roots of the eigenvalues of AAH.) Because it is desirable to be able to compute these quantities in real time, an efficient technique for computing the SVD is vital. In addition, emerging multicore processors like graphical processing units (GPUs) are bringing parallel processing capabilities to an ever increasing number of users. Since the computational tasks involved in array signal processing are well suited for parallelization, it is expected that these computations will be implemented using GPUs as soon as users have the necessary computational tools available to them. Thus, it is important to have an SVD algorithm that is suitable for these processors. This work explores the effectiveness of two different parallel SVD implementations on an NVIDIA Tesla C2050 GPU (14 multiprocessors, 32 cores per multiprocessor, 1.15 GHz clock - peed). The first algorithm is based on a two-step algorithm which bidiagonalizes the matrix using Householder transformations, and then diagonalizes the intermediate bidiagonal matrix through implicit QR shifts. This is similar to that implemented for real matrices by Lahabar and Narayanan ("Singular Value Decomposition on GPU using CUDA", IEEE International Parallel Distributed Processing Symposium 2009). The implementation is done in a hybrid manner, with the bidiagonalization stage done using the GPU while the diagonalization stage is done using the CPU, with the GPU used to update the U and V matrices. The second algorithm is based on a one-sided Jacobi scheme utilizing a sequence of pair-wise column orthogonalizations such that A is replaced by AV until the resulting matrix is sufficiently orthogonal (that is, equal to U ). V is obtained from the sequence of orthogonalizations, while can be found from the square root of the diagonal elements of AH A and, once is known, U can be found from column scaling the resulting matrix. These implementations utilize CUDA Fortran and NVIDIA's CUB LAS library. The primary goal of this study is to quantify the comparative performance of these two techniques against themselves and other standard implementations (for example, MATLAB). Considering that there is significant overhead associated with transferring data to the GPU and with synchronization between the GPU and the host CPU, it is also important to understand when it is worthwhile to use the GPU in terms of the matrix size and number of concurrent SVDs to be calculated.

99 GENERAL AND MISCELLANEOUS↗

Parallelization of a Proxy Transport App Using ComputeCPP and SYCL

This document details the attempts to parallelize a multi-group Monte Carlo proxy app using SYCL and the ComputeCPP Community Edition compiler. SYCL is a standard set by the Khronos Group and is described as a, “cross-platform abstraction layer that builds on the underlying concepts, portability and efficiency of OpenCL that enables code for heterogeneous processors to be written in a “single-source” style using completely standard C++.” ComputeCpp Community Edition (CE) is a heterogeneous parallel programming platform that provides a conformant implementation of the SYCL™ 1.2.1 Khronos specification. ComputeCPP provides a compiler and an SDK containing a build system to improve the usability of the compiler. The supported OpenCL 1.2 platforms for ComputeCpp are AMD® and Intel®, with experimental support for PTX which is used on Nvidia graphics cards as well as experimental support for ARM chips. All testing reported in this document was done on the Darwin computer cluster using an Nvidia GTX Titan X graphics card. Since ComputeCPP does not support using this card with OpenCL, ComputeCPP’s experimental PTX64 bit-code generation was used to generate device code.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Efficient reversible uniform and non-uniform random number generation in UNU.RAN

Reversible random number generations are useful in large-scale fault-tolerant parallel computations and parallel discrete event simulations that are based on reversible computation. The Universal Non-Uniform Random Number Generator (UNU.RAN) is one of the popular random number generators used in the simulation community, but the generators are forward-only in nature. In this paper, we develop new reverse algorithm for the default uniform random number generator algorithm of UNU.RAN and also a few nonuniform random generators that use the Transform Density Reduction (TDR) method. We verify the correctness of reversals of our algorithms and also provide performance results to demonstrate reverse computing runtime adds little overheads relative to its forward counterpart.

Yoginath, Srikanth↗