Search NASA⌕ Search

SEARCH · Search NASA

Results for “distributed parallelization”

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 199 records · Page 11

A New Capability of E4D For 3D Parallel Joint Inversion of DC Resistivity And Traveltime Data on Unstructured Mesh

A major challenge in interpreting geophysical data is how to derive consistent three-dimensional (3D) earth models of different physical properties from spatially and temporally limited measurements. Joint inversion with cross-gradient constraints is an approach to find such models by imposing structural similarities between different physical parameters. We have developed a parallel distributed-memory joint inversion code for direct-current (DC) resistivity and traveltime data using the cross-gradient constraint on unstructured mesh. The code utilizes existing E4D framework for parallel forward simulation, distributed storage and computation of the Jacobian matrix of forward operator, and parallel execution of matrix-vector multiplication during inversion. Besides, the joint inversion is solved by nonlinear conjugate gradient algorithm parallelized for DC resistivity and traveltime data. The joint inversion capability of E4D was tested using synthetic data from cross-borehole DC resistivity and traveltime data. The results indicate that the shape and size of the anomalies from the joint inversion are more reliable than those from separate inversions.

58 GEOSCIENCES↗

Formation of electron kappa distributions due to interactions with parallel propagating whistler waves

In space plasmas, charged particles are frequently observed to possess a high-energy tail, which is often modeled by a kappa-type distribution function. In this work, the formation of the electron kappa distribution in generation of parallel propagating whistler waves is investigated using fully nonlinear particle-in-cell (PIC) simulations. A previous research concluded that the bi-Maxwellian character of electron distributions is preserved in PIC simulations. We now demonstrate that for interactions between electrons and parallel propagating whistler waves, a non-Maxwellian high-energy tail can be formed, and a kappa distribution can be used to fit the electron distribution in time-asymptotic limit. The κ-parameter is found to decrease with increasing initial temperature anisotropy or decreasing ratio of electron plasma frequency to cyclotron frequency. The results might be helpful to understanding the origin of electron kappa distributions observed in space plasmas.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Final Technical Report

The nonlinear physics of electron magnetohydrodynamics (EMHD) in plasmas. Time-varying wave magnetic field exceeding the background magnetic field produces highly nonlinear whistler mode since the wave dispersion depends on the total magnetic field. There exists no theory for such whistler modes. The present experimental work is the first one to explore this regime of nonlinear whistlers. A field-reversed configuration has been found which has the same vortex topology as an MHD spheromak, termed a whistler spheromak. Whistler mirrors have compressed and twisted field lines propagating in the whistler mode. Their helicity properties have been studied. Whistler spheromaks and mirrors have different propagation and damping characteristics. Wave collisions have been studied. Head-on collisions of two whistler spheromaks form a stationary field-reversed configuration (FRC) without helicity. When whistler spheromaks are excited the toroidal current flows mainly in the toroidal null line. It is only carried by electrons since ion currents and displacement currents are negligible. A change in the poloidal (axial) magnetic field induces a toroidal electric field which drives the current. Magnetic energy is dissipated and converted into electron kinetic energy. This process is called magnetic reconnection in 2D geometries, which are simplifications for theoretical convenience but rarely occur in nature. A crucial aspect of reconnection is its rate, determined by the electron collisionality. Regular Coulomb collisions can rarely account for the observed reconnection rates. In the present experiments we have also observed fast reconnection and explained it by electron transit time damping in the finite-size null layer. Electrons move faster than a whistler spheromak, hence transit through the toroidal null line where they are freely accelerated. The transit time is essentially the collision time but no particle collisions are required. Strong electron heating and visible light emissions are only observed in whistler spheromaks and FRCs but not in mirrors or asymmetric configurations lacking magnetic null lines. The collisionless electron energization in a toroidal null line usually produces non-Maxwellian distributions. Off the null axis electrons gain more perpendicular than parallel energy. Distributions with T{sub {perpendicular}} > T{sub {parallel}} lead to whistler instabilities which have been observed. A whistler spheromak is a source of high-frequency whistler emissions. These are usually small amplitude whistlers propagating in a complicated background magnetic field. The waves are emitted from a moving source. High frequency whistlers propagate faster than the spheromak, thus partly move ahead of it and partly in the reverse direction. In test wave experiments wave growth opposite to the direction of the hot electron flow has been observed, confirming that Doppler-shifted cyclotron resonance instabilities account for the emission process. Propagating whistler mirrors produce no significant instabilities except when they interact with other fields which exhibit null lines. For example, a whistler mirror has been launched against a stationary FRC, resulting in strong FRC heating and whistler instabilities. In the whistler mirror configuration the antenna near-zone field produces a toroidal null line outside the coil which can also become a source for whistler emissions. Finally, nonlinear EMHD research has been extended to initially unmagnetized plasmas where a new nonlinear skin depth has been discovered. When a small-amplitude oscillating magnetic field is applied to a plasma the field penetration is governed by the skin depth, collisional or collisionless depending on frequency, collision frequency and plasma frequency. However, when the magnetic field increases the electrons become magnetized and the field penetration occurs in the whistler mode if the cyclotron frequency exceeds the oscillating frequency. This phenomenon has been observed. A loop antenna creates a dipole field which is frozen into the plasma for a half cycle and becomes the background field for the wave launched by the next half cycle. The field topology consists of field-reversed dipoles of decreasing strength with distance. The propagation region depends on field amplitude but not on the skin depth. Our research has been published in 13 scientific papers.

43 PARTICLE ACCELERATORS↗

Dynamical effects in proton breakup from exotic nuclei

This contribution discusses dynamical effects in proton breakup from a weakly bound state in an exotic nucleus on a heavy target. The Coulomb interactions between the proton and the core and the proton and the target are treated to all orders, including also the full multipole expansion of the Coulomb potential. The dynamics of proton Coulomb breakup is compared to that of an equivalent neutron of larger binding energy in order to elucidate the differences with the well understood neutron breakup mechanism. A number of experimentally measurable observables such as parallel momentum distributions, proton angular distributions and total breakup cross sections can be calculated. With respect to nuclear breakup it is found that a proton behaves exactly as a neutron of larger binding energy. The extra 'effective energy' is due to the combined core-target Coulomb barrier. In Coulomb breakup we distinguish the effect of the core-target Coulomb potential (called recoil effect), with respect to which the proton behaves again as a more bound neutron, from the direct proton-target Coulomb potential. The latter gives cross sections about an order of magnitude larger than the recoil term. The two effects give rise to complicated interferences in the parallel momentum distributions. They are instead easily separable in the proton angular distributions which are therefore suggested as a very useful observable for future experimental studies.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Dynamical effects in proton breakup from exotic nuclei

We study dynamical effects in proton breakup from a weakly bound state in an exotic nucleus on a heavy target. The Coulomb interactions between the proton and the core and between the proton and the target are treated to all orders, including also the full multipole expansion of the Coulomb potential. The dynamics of proton nuclear and Coulomb breakup is compared to that of an equivalent neutron of higher binding energy in order to elucidate the differences with the well-understood neutron breakup mechanism. A number of experimentally measurable observables such as parallel momentum distributions, proton angular distributions, and total breakup cross sections are calculated. With respect to nuclear breakup it is found that a proton behaves exactly as a neutron of higher binding energy. The extra ''effective energy'' is caused by the combined core-target Coulomb barrier. In Coulomb breakup we distinguish the effect of the core-target Coulomb potential (called the recoil effect), with respect to which the proton behaves, again, as a more bound neutron, from the direct proton-target Coulomb potential. The latter gives cross sections about an order of magnitude larger than the recoil term. The two effects give rise to complicated interferences in the parallel momentum distributions. They are instead easily separable in the proton angular distributions, which are therefore suggested as a very useful observable for future experimental studies.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Avoiding Communication in Primal and Dual Block Coordinate Descent Methods

Primal and dual block coordinate descent methods are iterative methods for solving regularized and unregularized optimization problems. Distributed-memory parallel implementations of these methods have become popular in analyzing large machine learning datasets. However, existing implementations communicate at every iteration, which, on modern data center and supercomputing architectures, often dominates the cost of floating-point computation. Recent results on communication-avoiding Krylov subspace methods suggest that large speedups are possible by re-organizing iterative algorithms to avoid communication. We show how applying similar algorithmic transformations can lead to primal and dual block coordinate descent methods that only communicate every $s$ iterations---where $s$ is a tuning parameter---instead of every iteration for the regularized least-squares problem. We show that the communication-avoiding variants reduce the number of synchronizations by a factor of $s$ on distributed-memory parallel machines without altering the convergence rate and attain strong scaling speedups of up to $6.1\times$ over the “standard algorithm" on a Cray XC30 supercomputer.

97 MATHEMATICS AND COMPUTING↗

Distribution of air-water mixtures in parallel vertical channels as an effect of the header geometry

Uneven phase distribution in heat exchangers is a cause of severe reductions in thermal performances of refrigeration equipment. To date, no general design rules are available to avoid phase separation in manifolds with several outlet channels, and even predicting the phase and mass distribution in parallel channels is a demanding task. In the present paper, measurements of two-phase air-water distributions are reported with reference to a horizontal header supplying 16 vertical upward channels. The effects of the operating conditions, the header geometry and the inlet port nozzle were investigated in the ranges of liquid and gas superficial velocities of 0.2-1.2 and 1.5-16.5 m/s, respectively. Among the fitting devices used, the insertion of a co-axial, multi-hole distributor inside the header confirmed the possibility of greatly improving the liquid and gas flow distribution by the proper selection of position, diameter and number of the flow openings between the supplying distributor and the system of parallel channels connected to the header. (author)

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Numerically stable, scalable formulas for parallel and online computation of higher-order multivariate central moments with arbitrary weights

Formulas for incremental or parallel computation of second order central moments have long been known, and recent extensions of these formulas to univariate and multivariate moments of arbitrary order have been developed. Such formulas are of key importance in scenarios where incremental results are required and in parallel and distributed systems where communication costs are high. We survey these recent results, and improve them with arbitrary-order, numerically stable one-pass formulas which we further extend with weighted and compound variants. We also develop a generalized correction factor for standard two-pass algorithms that enables the maintenance of accuracy over nearly the full representable range of the input, avoiding the need for extended-precision arithmetic. We then empirically examine algorithm correctness for pairwise update formulas up to order four as well as condition number and relative error bounds for eight different central moment formulas, each up to degree six, to address the trade-offs between numerical accuracy and speed of the various algorithms. Finally, we demonstrate the use of the most elaborate among the above mentioned formulas, with the utilization of the compound moments for a practical large-scale scientific application.

97 MATHEMATICS AND COMPUTING↗

Parallel Breadth-First Search on Distributed Memory Systems

Data-intensive, graph-based computations are pervasive in several scientific applications, and are known to be quite challenging to implement on distributed memory systems. In this work, we explore the design space of parallel algorithms for Breadth-First Search (BFS), a key subroutine in several graph algorithms. We present two highly-tuned parallel approaches for BFS on large parallel systems: a level-synchronous strategy that relies on a simple vertex-based partitioning of the graph, and a two-dimensional sparse matrix-partitioning-based approach that mitigates parallel communication overhead. For both approaches, we also present hybrid versions with intra-node multithreading. Our novel hybrid two-dimensional algorithm reduces communication times by up to a factor of 3.5, relative to a common vertex based approach. Our experimental study identifies execution regimes in which these approaches will be competitive, and we demonstrate extremely high performance on leading distributed-memory parallel systems. For instance, for a 40,000-core parallel execution on Hopper, an AMD Magny-Cours based system, we achieve a BFS performance rate of 17.8 billion edge visits per second on an undirected graph of 4.3 billion vertices and 68.7 billion edges with skewed degree distribution.

97 MATHEMATICS AND COMPUTING↗

A Massively Parallel Adaptive Fast Multipole Method on Heterogeneous Architectures

We describe a parallel fast multipole method (FMM) for highly nonuniform distributions of particles. We employ both distributed memory parallelism (via MPI) and shared memory parallelism (via OpenMP and GPU acceleration) to rapidly evaluate two-body nonoscillatory potentials in three dimensions on heterogeneous high performance computing architectures. We have performed scalability tests with up to 30 billion particles on 196,608 cores on the AMD/CRAY-based Jaguar system at ORNL. On a GPU-enabled system (NSF's Keeneland at Georgia Tech/ORNL), we observed 30x speedup over a single core CPU and 7x speedup over a multicore CPU implementation. By combining GPUs with MPI, we achieve less than 10 ns/particle and six digits of accuracy for a run with 48 million nonuniformly distributed particles on 192 GPUs.

99 GENERAL AND MISCELLANEOUS↗

An exploration of online-simulation-driven portfolio scheduling in Workflow Management Systems

Workflow Management Systems used to automate the execution of scientific workflow applications on parallel and distributed computing platforms must make scheduling decisions at runtime. A large number of workflow scheduling algorithms have been proposed in the literature, but often these algorithms are evaluated based on simplifying assumptions that may not hold in practice. Furthermore, published algorithm evaluation and/or comparison results are necessarily only for a subset of all possible scenarios, and thus may not include scenarios relevant to particular use-cases. Consequently, it is difficult for Workflow Management Systems (WMSs) developers to decide which scheduling algorithm should be implemented. To obviate this difficulty, one possible approach is to implement a portfolio of scheduling algorithms and select the most effective algorithm at runtime. One method for performing this selection is to run an online simulation for each algorithm in the portfolio. The algorithm that leads to the best performance, in simulation, is selected for future use. The above simulation-driven portfolio scheduling (SDPS) approach has been proposed in a few parallel and distributed computing contexts. The main objective of this work is to evaluate the feasibility and potential merit of SDPS if implemented in WMSs. Here we perform this evaluation using simulated WMS executions, where the simulations are instantiated from real-world platform and workflow configurations. Our main finding is that SDPS is on par with or outperforms an approach in which a single algorithm is used, where this algorithm is the one that performs best on average across all our experimental scenarios. Furthermore, we find that SDPS remains an attractive proposition even in the presence of high levels of simulation error and for simulators with relatively low levels of sophistication. In many of our experimental scenarios we find that mitigating simulation error at runtime can further improve performance. Finally, we show that simulation overhead can be made sufficiently low for SDPS to be feasible in practice.

97 MATHEMATICS AND COMPUTING↗

Knockout beyond the dripline

In this contribution, we will describe neutron and proton removal from {sup 9}C and {sup 7}Be which are two particularly interesting nuclei entering the nucleo-synthesis pp-chain [1, 2]. Neutron and proton removal reactions have been used in the past twenty years to probe the single-particle structure of exotic nuclei. The core parallel-momentum distribution can give information on the angular momentum and spin of the nucleon initial state while the total removal cross section is sensitive to the asymptotic part of the initial wave function and also to the reaction mechanism. Because knockout is a peripheral reaction from which the Asymptotic Normalization Constant (ANC) of the single-particle wave function can be extracted, it has been used as an indirect method to obtain the rate of reactions like {sup 8}B(p,γ){sup 9}C or {sup 7}Be(p,γ){sup 8}B. Nucleon removal has recently been applied by the HiRA collaboration [3] to situations in which the remaining “core” is beyond the drip line, such as {sup 8}C and {sup 6}Be, unbound by one or more protons, and whose excitation-energy spectrum can be obtained by the invariant-mass method. By gating on the ground-state peak, “core” parallel-momentum distributions and total knockout cross sections have been obtained similar to previous studies with well-bound “cores”. In addition for each projectile, knock out to final bound states has also been obtained in several cases. We will report on the theoretical description and comparison to this experimental data for a few cases for which advances in the accuracy of the transfer-to-the continuum model [4, 5] have been made [6]. These include the use, when available, of “ab-initio” overlaps for the initial state [7] and in particular their ANC values [8]. Also, the construction of a nucleus-target folding potential for the treatment of the core-target S-matrix [9] using for the cores “ab-initio” densities [10] and state-of-the-art n−{sup 9}Be optical potentials [11]. Preliminary results and open problems will be discussed.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

Performance Evaluation of Different Parallel Programming Models in SCALE-Shift Sequences for Criticality and Shielding Applications [Abstract]

The SCALE code system has been widely used for nuclear criticality safety, reactor physics, radiation shielding, source term generation, and inventory analyses by researchers, industry, and regulatory bodies. Although limited support for shared- and distributed-memory parallel processing was introduced via C++ threading, OpenMP, and MPI, a hybrid parallel programming model with both distributed- and shared-memory parallelism has not been fully supported in the SCALE code system.

Nuclear Criticality Safety Program (NCSP)↗

Parallel Hybrid Metaheuristics with Distributed Intensification and Diversification for Large-scale Optimization in Big Data Statistical Analysis

Important insights into many problems that are traditionally analyzed via statistical models can be obtained by re-formulating and evaluating within a large-scale optimization framework. The theoretical underpinnings of the statistical model often shift the goal of the solution space traversal from a traditional search for a single optimal solution to a traversal with the purpose of yielding a set of high quality, independent solutions. We examine statistical frameworks with astronomical solution spaces where the independence requirement constitutes a significant additional challenge for standard optimization methodologies. We design a hybrid metaheuristic with specialized intensification and diversification protocols in the base search algorithm. We extend our algorithm to the high-performance-computing realm using the Stampede2 supercomputer. We experimentally demonstrate the effectiveness of our algorithm to utilize multiple processors to collaboratively hill climb, broadcast messages to one another regarding landscape characteristics, diversify across the solution landscape, and request aid in climbing particularly difficult peaks.

Cho, Wendy K. Tam↗

A Parallel Ghosting Algorithm for The Flexible Distributed Mesh Database

Critical to the scalability of parallel adaptive simulations are parallel control functions including load balancing, reduced inter-process communication and optimal data decomposition. In distributed meshes, many mesh-based applications frequently access neighborhood information for computational purposes which must be transmitted efficiently to avoid parallel performance degradation when the neighbors are on different processors. This article presents a parallel algorithm of creating and deleting data copies, referred to as ghost copies, which localize neighborhood data for computation purposes while minimizing inter-process communication. The key characteristics of the algorithm are: (1) It can create ghost copies of any permissible topological order in a 1D, 2D or 3D mesh based on selected adjacencies. (2) It exploits neighborhood communication patterns during the ghost creation process thus eliminating all-to-all communication. (3) For applications that need neighbors of neighbors, the algorithm can create n number of ghost layers up to a point where the whole partitioned mesh can be ghosted. Strong and weak scaling results are presented for the IBM BG/P and Cray XE6 architectures up to a core count of 32,768 processors. The algorithm also leads to scalable results when used in a parallel super-convergent patch recovery error estimator, an application that frequently accesses neighborhood data to carry out computation.

Mubarak, Misbah↗

A multilingual programming model for coupled systems.

Multiphysics and multiscale simulation systems share a common software requirement-infrastructure to implement data exchanges between their constituent parts-often called the coupling problem. On distributed-memory parallel platforms, the coupling problem is complicated by the need to describe, transfer, and transform distributed data, known as the parallel coupling problem. Parallel coupling is emerging as a new grand challenge in computational science as scientists attempt to build multiscale and multiphysics systems on parallel platforms. An additional coupling problem in these systems is language interoperability between their constituent codes. We have created a multilingual parallel coupling programming model based on a successful open-source parallel coupling library, the Model Coupling Toolkit (MCT). This programming model's capabilities reach beyond MCT's native Fortran implementation to include bindings for the C++ and Python programming languages. We describe the method used to generate the interlanguage bindings. This approach enables an object-based programming model for implementing parallel couplings in non-Fortran coupled systems and in systems with language heterogeneity. We describe the C++ and Python versions of the MCT programming model and provide short examples. We report preliminary performance results for the MCT interpolation benchmark. We describe a major Python application that uses the MCT Python bindings, a Python implementation of the control and coupling infrastructure for the community climate system model. We conclude with a discussion of the significance of this work to productivity computing in multidisciplinary computational science.

97 MATHEMATICS AND COMPUTING↗