SparTen: Leveraging Kokkos for On-node Parallelism in a Second-Order Method for Fitting Canonical Polyadic Tensor Models to Poisson Data.
Abstract not provided.
Teranishi, Keita↗
SEARCH · Search NASA
Results for “parallel methods”
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.
PLATO (Parallel Load Assignment Tool): A Parallel Workload Partitioner for Particle-based Methods
A key first step in many large-scale high-performance scientific computing applications is the distribution of the computational domain across all participating processors of a parallel computing platform such as a shared memory machine, cluster or a tightly-coupled supercomputer. The primary goal of such a partitioning step is to ensure that the computational work performed by each process remains load-balanced throughout the execution of the parallel application. In this project, a general-purpose parallel partitioning tool, code-named PLATO, was designed and implemented. PLATO assumes that the workload is a set of points in an n-dimensional metric space. These points are abstractions of computational units that may represent point-like computational units such as celestial objects in astrophysics, atoms in materials sciences and molecules in biology/chemistry, or may represent more abstract computational units such as grid points in finite difference methods, grid elements in finite element methods or n-dimensional points in the feature space of a host of other applications. PLATO supports several key and useful capabilities, namely: (a) multiple load partitioning strategies, (b) support for arbitrary user-defined processor topology, (c) pre-fetching data from neighboring processors based on user-defined parameters, and (d) support for arbitrary number and types of attributes associated with each point in the computational domain. PLATO is implemented in the C programming language using Message Passing Interface (MPI) for inter-processor communications and its parallel performance tested on up to 48 processors of a shared memory machine. This report includes a detailed performance study of PLATO using parallel speedup and iso-efficiency as the two main metrics of parallel performance. In addition, a user manual and DOXYGEN generated code documentation have been included for completeness.
97 MATHEMATICS AND COMPUTING↗
Communication-avoiding optimization methods for distributed massive-scale sparse inverse covariance estimation
Across a variety of scientific disciplines, sparse inverse covariance estimation is a popular tool for capturing the underlying dependency relationships in multivariate data. Unfortunately, most estimators are not scalable enough to handle the sizes of modern high-dimensional data sets (often on the order of terabytes), and assume Gaussian samples. To address these deficiencies, we introduce HP-CONCORD, a highly scalable optimization method for estimating a sparse inverse covariance matrix based on a regularized pseudolikelihood framework, without assuming Gaussianity. Our parallel proximal gradient method uses a novel communication-avoiding linear algebra algorithm and runs across a multi-node cluster with up to 1k nodes (24k cores), achieving parallel scalability on problems with up to ≈819 billion parameters (1.28 million dimensions); even on a single node, HP-CONCORD demonstrates scalability, outperforming a state-of-the-art method. We also use HP-CONCORD to estimate the underlying dependency structure of the brain from fMRI data, and use the result to identify functional regions automatically. The results show good agreement with a clustering from the neuroscience literature.
97 MATHEMATICS AND COMPUTING↗
Communication-Avoiding Optimization Methods for Distributed Massive-Scale Sparse Inverse Covariance Estimation.
Across a variety of scientific disciplines, sparse inverse covariance estimation is a popular tool for capturing the underlying dependency relationships in multivariate data. Unfortunately, most estimators are not scalable enough to handle the sizes of modern high-dimensional data sets (often on the order of terabytes), and assume Gaussian samples. To address these deficiencies, we introduce HP-CONCORD, a highly scalable optimization method for estimating a sparse inverse covariance matrix based on a regularized pseudolikelihood framework, without assuming Gaussianity. Our parallel proximal gradient method uses a novel communication-avoiding linear algebra algorithm and runs across a multi-node cluster with up to 1k nodes (24k cores), achieving parallel scalability on problems with up to ~819 billion parameters (1.28 million dimensions); even on a single node, HP-CONCORD demonstrates scalability, outperforming a state-of-the-art method. We also use HP-CONCORD to estimate the underlying dependency structure of the brain from fMRI data, and use the result to identify functional regions automatically. The results show good agreement with a clustering from the neuroscience literature.
Koanantakool, Penporn↗
Cooperative storage of shared files in a parallel computing system with dynamic block size
Improved techniques are provided for parallel writing of data to a shared object in a parallel computing system. A method is provided for storing data generated by a plurality of parallel processes to a shared object in a parallel computing system. The method is performed by at least one of the processes and comprises: dynamically determining a block size for storing the data; exchanging a determined amount of the data with at least one additional process to achieve a block of the data having the dynamically determined block size; and writing the block of the data having the dynamically determined block size to a file system. The determined block size comprises, e.g., a total amount of the data to be stored divided by the number of parallel processes. The file system comprises, for example, a log structured virtual parallel file system, such as a Parallel Log-Structured File System (PLFS).
97 MATHEMATICS AND COMPUTING↗
Lattice Quantum Chromodynamics with Overlap Fermions on GPUs
Lattice quantum chromodynamics (QCD) calculations were one of the first applications to demonstrate the potential of GPUs in the area of high-performance computing; the nature of lattice QCD calculations matches well with the GPU's computational model. This article discusses ways to effectively use GPUs for lattice calculations using the overlap operator, a discretization that preserves chiral symmetry even at nonzero lattice spacing and makes possible lattice QCD simulations in the parameter region relevant to nuclear physics. The author shows that the large memory footprint of these codes requires the use of multiple GPUs in parallel and discusses methods for implementing this operator efficiently: hybrid CPU/GPU memory use for eigensolvers and MPI/OpenMP/CUDA parallelization strategies required to take full advantage of both GPU and CPU resources. He then compares the performance of codes on a GPU cluster and a CPU cluster with similar interconnects, discussing the strong scaling for problem sizes relevant to current lattice QCD simulations.
Computer Science↗
Heterodyne frequency-domain multispectral diffuse optical tomography of breast cancer in the parallel-plane transmission geometry
Purpose: The authors introduce a state-of-the-art all-optical clinical diffuse optical tomography (DOT) imaging instrument which collects spatially dense, multispectral, frequency-domain breast data in the parallel-plate geometry. Methods: The instrument utilizes a CCD-based heterodyne detection scheme that permits massively parallel detection of diffuse photon density wave amplitude and phase for a large number of source–detector pairs (10{sup 6}). The stand-alone clinical DOT instrument thus offers high spatial resolution with reduced crosstalk between absorption and scattering. Other novel features include a fringe profilometry system for breast boundary segmentation, real-time data normalization, and a patient bed design which permits both axial and sagittal breast measurements. Results: The authors validated the instrument using tissue simulating phantoms with two different chromophore-containing targets and one scattering target. The authors also demonstrated the instrument in a case study breast cancer patient; the reconstructed 3D image of endogenous chromophores and scattering gave tumor localization in agreement with MRI. Conclusions: Imaging with a novel parallel-plate DOT breast imager that employs highly parallel, high-resolution CCD detection in the frequency-domain was demonstrated.
60 APPLIED LIFE SCIENCES↗
Massively parallel quantum chemical density matrix renormalization group method
Here, we present, to the best of our knowledge, the first attempt to exploit the supercomputer platform for quantum chemical density matrix renormalization group (QCDMRG) calculations. We have developed the parallel scheme based on the in-house MPI global memory library, which combines operator and symmetry sector parallelisms, and tested its performance on three different molecules, all typical candidates for QC-DMRG calculations. In case of the largest calculation, which is the nitrogenase FeMo cofactor cluster with the active space comprising 113 electrons in 76 orbitals and bond dimension equal to 6000, our parallel approach scales up to approximately 2000 CPU cores.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
Large Scale Tensor Factorization via Parallel Sketches
Tensor factorization methods have recently gained increased popularity. A key feature that renders tensors attractive is the ability to directly model multi-relational data. In this work, we propose ParaSketch, a parallel tensor factorization algorithm that enables massive parallelism, to deal with large tensors. The idea is to compress the large tensor into multiple small tensors, decompose each small tensor in parallel, and combine the results to reconstruct the desired latent factors. Prior art in this direction entails potentially very high complexity in the (Gaussian) compression and final combining stages. Adopting sketching matrices for compression, the proposed method enjoys a dramatic reduction in compression complexity, and features a much lighter combining step. Moreover, theoretical analysis shows that the compressed tensors inherit latent identifiability under mild conditions, hence establishing correctness of the overall approach. Numerical experiments corroborate the theory and demonstrate the effectiveness of the proposed algorithm.
block term decomposition↗
Valence : A Massively Parallel Implementation of the Variational Subspace Valence Bond Method
This work describes the software package, Valence, for the calculation of molecular energies using the variational subspace valence bond (VSVB) method. VSVB is an ab initio electronic structure method based on nonorthogonal orbitals. Important features of practical value include high parallel scalability, wave functions that can be constructed automatically by combining orbitals from previous calculations, and ground and excited states that can be modeled with a single configuration or determinant. Here, the open-source software package includes tools to generate wave functions, a database of generic orbitals, example input files, and a library build intended for integration with other packages. We also present the interface to an external software package, enabling the computation of optimized molecular geometries and vibrational frequencies.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
A real-space Green’s function method for thenumerical solution of Maxwell’s equations
A new method for solving the transverse part of the free-space Maxwell equations in three dimensions is presented. By taking the Helmholtz decomposition of the electric field and current sources and considering only the divergence-free parts, we obtain an explicit real-space representation for the transverse propagator that explicitly respects finite speed of propagation. Because the propagator involves convolution against a singular distribution, we regularize via convolution with smoothing kernels (B-splines) prior to sampling based on a method due to Beyer and LeVeque (1992). We show that the ultimate discrete convolutional propagator can be constructed to attain an arbitrarily high order of accuracy by using higher-order regularizing kernels and finite difference stencils and that it satisfies von Neumann’s stability condition. Furthermore, the propagator is compactly supported and can be applied using Hockney’s method (1970) and parallelized using the same observation as made by Vay, Haber, and Godfrey (2013), leading to a method that is computationally efficient.
Lo, Boris↗
Parallel multiple-chain DRAM MCMC for large-scale geosteering inversion and uncertainty quantification
Geosteering is the proactive control of a wellbore placement based on the downhole measurements, aiming at maximizing economic production from the well. We report the recent development of azimuthal resistivity logging-whiledrilling (LWD) tools have a much larger depth of detection, thus sets higher demands for geosteering inversion as more unknown parameters need to be taken into account in the earth model to be inverted. For complicated nonlinear problems, traditional deterministic inversion methods are more likely to be trapped near local optima. In general, Markov chain Monte Carlo (MCMC) inversion methods are more capable of finding the global optimal solution and providing additional uncertainty analyses by sampling from the target distribution. However, MCMC methods usually have the problem of slow convergence. Even though optimization methods like delayed rejection (DR) and adaptive Metropolis (AM) were proposed to speed up the convergence, using MCMC methods to solve geosteering inversion problems may still incur unacceptable time cost. To reduce the sampling cost of MCMC methods, we use parallel multiple-chain DRAM MCMC methods to solve geosteering inverse problems and estimate the corresponding uncertainty. A clustering method, density-based spatial clustering of applications with noise (DBSCAN), is applied to select the best solution among many results. Lastly, the simulation results show that running many relatively short MCMC chains for one problem can obtain a similar result as running a long single chain. Besides, avoiding communications between multiple Markov chains during the sampling process yields almost linear scalability.
97 MATHEMATICS AND COMPUTING↗
Mechanisms of interfacial reactivity in near surface and extreme environments
The local water structure surrounding ions in aqueous solutions greatly affects their chemical properties such as reaction rates, ion association, and proton and electron transport. These properties result in the behavior of ions in natural aqueous environments. For example ore transport is facilitated by chloride ion pair formation and the reaction of ions in an interface is strongly dependent on the dehydration of the ion hydration shell. We are developing the use of high-resolution XAFS observations and 1st principles based MD-XAFS analysis (spectra simulated using 1st principle methods with no adjustable parameters, AIMD) to interpret the solution properties of strongly interacting aqueous solutes under arbitrary pressure and temperature conditions. In the 1st principle MD-XAFS method density functional theory (DFT) based MD simulations(Car and Parrinello, 1985) are used to generate a large ensemble of structural snap shots of the hydration region. These are then used to generate scattering intensities. I emphasize three points about this novel approach to analyzing XAFS data. 1st: As illustrated in Figure 1, the level of agreement between the calculated and observed intensities is considerably higher than has been obtained by any XAFS analysis to date (note 2nd shell region, R> 2 Å). 2nd: This result was obtained from a parameter free simulation with no fitting of the interaction potentials to any data. This supports the use of these methods for more difficult environments and more complex solutes (polyions). 3rd: New information about the shell structure (Figure 1) is now available because of this more detailed agreement. We note also that both multiple scattering and second shell features are well represented in the analysis. As far as we know this is the 1st analysis of second shell structure and multiple scattering. Excellent agreement has been obtained for most of the third row metal ions: Ca 2+ , Zn 2+ , Cu 2+ , Ni 2+ , Co 2+ , Mn 2+ , Fe 3+ , Cr 3+ . Calculations on these systems are demanding because of their open electronic shells, and high ionic charge. Principal Investigator: Professor John Weare (University of California, San Diego) The prediction of the interactions of geochemical fluids with minerals, nanoparticles, and colloids under extreme near surface conditions of temperature (T) and pressure (P) is a grand challenge research need in geosciences (U.S. DOE 2007, Basic Research Needs for Geosciences: Facilitating the 21st Energy Systems.). To evaluate the impact of these processes on energy production and management strategies it is necessary to have a high level of understanding of the interaction between complex natural fluids and mineral formations. This program emphasizes 1st principle parameter free simulations of complex chemical processes in solutions, in the mineral phase, and in the interfaces between these phases The development of new computational tools (with emphasis on oxide materials and reaction dynamics) tailored to treat wide range of conditions and time scales experienced in such geochemical applications is has been developed. Because of the sensitivity of the interaction in these systems to electronic structure and local bonding environments, and of the need to describe bond breaking/formation, our simulations are based on interactions calculated at the electronic structure level (ab-initio molecular dynamics, AIMD). The progress in the computational aspects of program may be summarized in terms of the following themes (objectives); Development of efficient parameter free dynamical simulation technology based on 1st principles force and energy calculations especially adapted for geochemical applications (e.g., mineral, interfaces and aqueous solutions) (continuing program); Calculation of the dynamics of water structure of in the surface-water interface of transition metal oxides and oxihydroxides; and Development of improved (beyond DFT+GGA) electronic structure calculations for minerals and the interface region that more accurately calculate electron correlation, spin density, and localization. The focus of the program is also on the iron oxide and oxihydroxide minerals and Fe 2+ (aq)/Fe 3+ (cr) oxidation in the mineral solution interface region. These methods included the development of model Hamilitonian methods that can be solved to near convergence for single site models (DMFT) and many-body perturbation methods (MP2, GW); Development of time decomposition methods to extend time scales of molecular dynamics (MD) simulations and support the use of high complexity electronic structure calculations (MP2, CCSD(T)) of forces for use in dynamical simulations where very high chemical accuracy is required (microsolvated reactions in absorbed surface layers); and The development of a new linear scaling finite element solver for eigenvalue problem that supports solution of quantum problems with unusual potential and boundary values. Application progress of the above new simulation technology to problems of geochemical interests includes; The prediction of metal oxide surface structure and the reduction/oxidation of Fe 3+ (cr)/Fe 2+ (aq) in metal oxide (hematite, goethite)/solution interfaces. Result: water interacts strongly with the 001 Hematite surface; interaction of water with the 100 goethite is weak; The study of ion solvation and the composition of ion hydration shells under extreme conditions (focus on Fe 3+/2+ , Al 3+ and Mg 2+ and their hydroxide speciation). Result: Ion association in water solutions can be calculated from 1st principle methods. Efficient sampling of the free energy requires more development; The continued development of new high resolution analysis of XAFS scattering of disordered systems (particularly Al, Mg) and of XANES calculations for aqueous ions. Result: EXAFS spectra can be calculated to high accuracy with DFT level dynamic simulations; The exploration of electron localization and electron transport in metal oxides (highly correlated materials). Result: Proper description of electron localization requires levels of calculation beyond DFT; and Localization of electrons in DFT type Hamiltonians was studied. Result: For very Dirac high exchange new solutions (New unphysical bifrucations) to the eigenvalue problem are found. The program was highly collaborative involving faculty and students in mathematics, physics and computer science departments as well as coworkers at the Pacific Northwest National Laboratories (PNNL). The students in this program had the opportunity to develop skills in the development of methods, the implementation of method on high performance parallel computers and the application of these methods to problem in geochemical science. Much of the software that was developed was incorporated in the NWchem software package maintained by PNNL.
58 GEOSCIENCES↗
Evaluating the Potential of a Laplacian Linear Solver.
A new approach for solving Laplacian linear systems proposed by Kelner et al. involves the random sampling and update of fundamental cycles in a graph. We evaluate the performance of this approach on a variety of real world graphs. We examine di erent ways to choose the set of cycles and their sequence of updates with the goal of providing more exibility and potential parallelism. We propose a parallel model of the Kelner et al. method for evaluating potential parallelism concerned with minimizing the span of edges updated at every iteration. We provide experimental results comparing the potential parallelism of the fundamental cycle basis and the extended basis. Our preliminary experiments show that choosing a non-fundamental set of cycles can save signi cant work compared to a fundamental cycle basis.
97 MATHEMATICS AND COMPUTING↗
Parallel algorithms for hyperdynamics and local hyperdynamics
Hyperdynamics (HD) is a method for accelerating the timescale of standard molecular dynamics (MD). It can be used for simulations of systems with an energy potential landscape that is a collection of basins, separated by barriers, where transitions between basins are infrequent. HD enables the system to escape from a basin more quickly while enabling a statistically accurate renormalization of the simulation time, thus effectively boosting the timescale of the simulation. In [Kim, Perez, Voter, J Chem Phys, 139:144110, 2013)1, a local version of HD was formulated, which exploits the intrinsic locality characteristic typical of most systems to mitigate the poor scaling properties of standard HD as the system size is increased. In this paper, we discuss how both HD and local HD can be formulated to run efficiently in parallel. We have implemented these ideas in the LAMMPS MD code, which means HD can be used with any interatomic potential LAMMPS supports. Together, these parallel methods allow simulations of any size to achieve the time acceleration offered by HD (which can be orders of magnitude), at a cost 3-5x that of standard MD. As examples, we performed two simulations of a million-atom system to model the diffusion and clustering of Pt adatoms on a large patch of Pt(100) surface for 80 and 160 μs.
74 ATOMIC AND MOLECULAR PHYSICS↗
Parallel macromolecular delivery and biochemical/electrochemical interface to cells employing nanostructures
Systems and methods are described for parallel macromolecular delivery and biochemical/electrochemical interface to whole cells employing carbon nanostructures including nanofibers and nanotubes. A method includes providing a first material on at least a first portion of a first surface of a first tip of a first elongated carbon nanostructure; providing a second material on at least a second portion of a second surface of a second tip of a second elongated carbon nanostructure, the second elongated carbon nanostructure coupled to, and substantially parallel to, the first elongated carbon nanostructure; and penetrating a boundary of a biological sample with at least one member selected from the group consisting of the first tip and the second tip.
59 BASIC BIOLOGICAL SCIENCES↗
Methods and apparatus for multi-resolution replication of files in a parallel computing system using semantic information
Techniques are provided for storing files in a parallel computing system using different resolutions. A method is provided for storing at least one file generated by a distributed application in a parallel computing system. The file comprises one or more of a complete file and a sub-file. The method comprises the steps of obtaining semantic information related to the file; generating a plurality of replicas of the file with different resolutions based on the semantic information; and storing the file and the plurality of replicas of the file in one or more storage nodes of the parallel computing system. The different resolutions comprise, for example, a variable number of bits and/or a different sub-set of data elements from the file. A plurality of the sub-files can be merged to reproduce the file.
97 MATHEMATICS AND COMPUTING↗
From NWChem to NWChemEx: Evolving with the Computational Chemistry Landscape
Since the advent of the first computers, chemists have been at the forefront of using computers to understand and solve complex chemical problems. As the hardware and software have evolved, so have the theoretical and computational chemistry methods and algorithms. Parallel computers clearly changed the common computing paradigm in the late 1970s and 80s, and the field has again seen a paradigm shift with the advent of graphical processing units. This review explores the challenges and some of the solutions in transforming software from the terascale to the petascale and now to the upcoming exascale computers. While discussing the field in general, NWChem and its redesign, NWChemEx, will be highlighted as one of the early codesign projects to take advantage of massively parallel computers and emerging software standards to enable large scientific challenges to be tackled.
37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗