Search NASA⌕ Search

SEARCH · Search NASA

Results for “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 55 records · Page 3

A Performance Model for Allocating the Parallelism in a Multigrid-in-Time Solver

The traditional way to numerically solve time-dependent problems is to sequentially march through time, solving for one time step and then the next. The parallelism in this approach is limited to the spatial dimension, which is quickly exhausted, causing the gain from using more processors to solve a problem to diminish. One approach to overcome this barrier is to use methods that are parallel in time. These methods have the potential to achieve dramatically better performance compared to time-stepping approaches, but achieving this performance requires carefully choosing the amount of parallelism devoted to space versus the amount devoted to time. Here, we present a performance model that, for a multigrid-in-time solver, makes the decision on when to switch to parallel-in-time and on how much parallelism to devote to space vs. time. In our experiments, the model selects the best parallel configuration in most of our test cases and a configuration close to the best one in all other cases.

97 MATHEMATICS AND COMPUTING↗

Accelerating Lattice QCD Multigrid on GPUs Using Fine-Grained Parallelization

The past decade has witnessed a dramatic acceleration of lattice quantum chromodynamics calculations in nuclear and particle physics. This has been due to both significant progress in accelerating the iterative linear solvers using multi-grid algorithms, and due to the throughput improvements brought by GPUs. Deploying hierarchical algorithms optimally on GPUs is non-trivial owing to the lack of parallelism on the coarse grids, and as such, these advances have not proved multiplicative. Using the QUDA library, we demonstrate that by exposing all sources of parallelism that the underlying stencil problem possesses, and through appropriate mapping of this parallelism to the GPU architecture, we can achieve high efficiency even for the coarsest of grids. Results are presented for the Wilson-Clover discretization, where we demonstrate up to 10x speedup over present state-of-the-art GPU-accelerated methods on Titan. Finally, we look to the future, and consider the software implications of our findings.

Clark, M. A.↗

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↗

Parallel pivoting combined with parallel reduction

Parallel algorithms for triangularization of large, sparse, and unsymmetric matrices are presented. The method combines the parallel reduction with a new parallel pivoting technique, control over generations of fill-ins and a check for numerical stability, all done in parallel with the work being distributed over the active processes. The parallel technique uses the compatibility relation between pivots to identify parallel pivot candidates and uses the Markowitz number of pivots to minimize fill-in. This technique is not a preordering of the sparse matrix and is applied dynamically as the decomposition proceeds.

Alaghband, Gita↗

Parallel pivoting combined with parallel reduction and fill-in control

Parallel algorithms for triangularization of large, sparse, and unsymmetric matrices are presented. The method combines the parallel reduction with a new parallel pivoting technique, control over generation of fill-ins and check for numerical stability, all done in parallel with the work being distributed over the active processes. The parallel pivoting technique uses the compatibility relation between pivots to identify parallel pivot candidates and uses the Markowitz number of pivots to minimize fill-in. This technique is not a preordering of the sparse matrix and is applied dynamically as the decomposition proceeds.

Alaghband, Gita↗

A fast algorithm for parallel computation of multibody dynamics on MIMD parallel architectures

In this paper the implementation of a parallel O(LogN) algorithm for computation of rigid multibody dynamics on a Hypercube MIMD parallel architecture is presented. To our knowledge, this is the first algorithm that achieves the time lower bound of O(LogN) by using an optimal number of O(N) processors. However, in addition to its theoretical significance, the algorithm is also highly efficient for practical implementation on commercially available MIMD parallel architectures due to its highly coarse grain size and simple communication and synchronization requirements. We present a multilevel parallel computation strategy for implementation of the algorithm on a Hypercube. This strategy allows the exploitation of parallelism at several computational levels as well as maximum overlapping of computation and communication to increase the performance of parallel computation.

Fijany, Amir↗

NAS Parallel Benchmark. Results 11-96: Performance Comparison of HPF and MPI Based NAS Parallel Benchmarks

High Performance Fortran (HPF), the high-level language for parallel Fortran programming, is based on Fortran 90. HALF was defined by an informal standards committee known as the High Performance Fortran Forum (HPFF) in 1993, and modeled on TMC's CM Fortran language. Several HPF features have since been incorporated into the draft ANSI/ISO Fortran 95, the next formal revision of the Fortran standard. HPF allows users to write a single parallel program that can execute on a serial machine, a shared-memory parallel machine, or a distributed-memory parallel machine. HPF eliminates the complex, error-prone task of explicitly specifying how, where, and when to pass messages between processors on distributed-memory machines, or when to synchronize processors on shared-memory machines. HPF is designed in a way that allows the programmer to code an application at a high level, and then selectively optimize portions of the code by dropping into message-passing or calling tuned library routines as 'extrinsics'. Compilers supporting High Performance Fortran features first appeared in late 1994 and early 1995 from Applied Parallel Research (APR) Digital Equipment Corporation, and The Portland Group (PGI). IBM introduced an HPF compiler for the IBM RS/6000 SP/2 in April of 1996. Over the past two years, these implementations have shown steady improvement in terms of both features and performance. The performance of various hardware/ programming model (HPF and MPI (message passing interface)) combinations will be compared, based on latest NAS (NASA Advanced Supercomputing) Parallel Benchmark (NPB) results, thus providing a cross-machine and cross-model comparison. Specifically, HPF based NPB results will be compared with MPI based NPB results to provide perspective on performance currently obtainable using HPF versus MPI or versus hand-tuned implementations such as those supplied by the hardware vendors. In addition we would also present NPB (Version 1.0) performance results for the following systems: DEC Alpha Server 8400 5/440, Fujitsu VPP Series (VX, VPP300, and VPP700), HP/Convex Exemplar SPP2000, IBM RS/6000 SP P2SC node (120 MHz) NEC SX-4/32, SGI/CRAY T3E, SGI Origin2000.

Saini, Subash↗

Shift-and-invert parallel spectral transformation eigensolver: Massively parallel performance for density-functional based tight-binding

The Shift-and-invert parallel spectral transformations (SIPs), a computational approach to solve sparse eigenvalue problems, is developed for massively parallel architectures with exceptional parallel scalability and robustness. The capabilities of SIPs are demonstrated by diagonalization of density-functional based tight-binding (DFTB) Hamiltonian and overlap matrices for single-wall metallic carbon nanotubes, diamond nanowires, and bulk diamond crystals. The largest (smallest) example studied is a 128,000 (2000) atom nanotube for which ~330,000 (~5600) eigenvalues and eigenfunctions are obtained in ~190 (~5) seconds when parallelized over 266,144 (16,384) Blue Gene/Q cores. Weak scaling and strong scaling of SIPs are analyzed and the performance of SIPs is compared with other novel methods. Different matrix ordering methods are investigated to reduce the cost of the factorization step, which dominates the time-to-solution at the strong scaling limit. As a result, a parallel implementation of assembling the density matrix from the distributed eigenvectors is demonstrated.

36 MATERIALS SCIENCE↗

Revisiting Parallel Cyclic Reduction and Parallel Prefix-Based Algorithms for Block Tridiagonal System of Equations

Simulations that require solutions of block tridiagonal systems of equations rely on fast parallel solvers for runtime efficiency. Leading parallel solvers that are highly effective for general systems of equations, dense or sparse, are limited in scalability when applied to block tridiagonal systems. This paper presents scalability results as well as detailed analyses of two parallel solvers that exploit the special structure of block tridiagonal matrices to deliver superior performance, often by orders of magnitude. A rigorous analysis of their relative parallel runtimes is shown to reveal the existence of a critical block size that separates the parameter space spanned by the number of block rows, the block size and the processor count, into distinct regions that favor one or the other of the two solvers. Dependence of this critical block size on the above parameters as well as on machine-specific constants is established. These formal insights are supported by empirical results on up to 2,048 cores of a Cray XT4 system. To the best of our knowledge, this is the highest reported scalability for parallel block tridiagonal solvers to date.

Seal, Sudip K↗

A Massively Parallel Implementation of the CCSD(T) Method Using the Resolution-of-the-Identity Approximation and a Hybrid Distributed/Shared Memory Parallelization Model

In this work, a parallel algorithm is described for the coupled-cluster singles and doubles method augmented with a perturbative correction for triple excitations [CCSD(T)] using the resolution-of-the-identity (RI) approximation for two-electron repulsion integrals (ERIs). The algorithm bypasses the storage of four-center ERIs by adopting an integral-direct strategy. The CCSD amplitude equations are given in a compact quasi-linear form by factorizing them in terms of amplitude-dressed three-center intermediates. A hybrid MPI/OpenMP parallelization scheme is employed, which uses the OpenMP-based shared memory model for intranode parallelization and the MPI-based distributed memory model for internode parallelization. Parallel efficiency has been optimized for all terms in the CCSD amplitude equations. Two different algorithms have been implemented for the rate-limiting terms in the CCSD amplitude equations that entail and -scaling computational costs, where N O and N V denote the number of correlated occupied and virtual orbitals, respectively. One of the algorithms assembles the four-center ERIs requiring N V 4 and N O 2 N V 2 -scaling memory costs in a distributed manner on a number of MPI ranks, while the other algorithm completely bypasses the assembling of quartic memory-scaling ERIs and thus largely reduces the memory demand. It is demonstrated that the former memory-expensive algorithm is faster on a few hundred cores, while the latter memory-economic algorithm shows a better strong scaling in the limit of a few thousand cores. The program is shown to exhibit a near-linear scaling, in particular for the compute-intensive triples correction step, on up to 8000 cores. The performance of the program is demonstrated via calculations involving molecules with 24–51 atoms and up to 1624 atomic basis functions. As the first application, the complete basis set (CBS) limit for the interaction energy of the π-stacked uracil dimer from the S66 data set has been investigated. This work reports the first calculation of the interaction energy at the CCSD(T)/aug-cc-pVQZ level without local orbital approximation. The CBS limit for the CCSD correlation contribution to the interaction energy was found to be -8.01 kcal/mol, which agrees very well with the value -7.99 kcal/mol reported by Schmitz, Hättig, and Tew [ Phys. Chem. Chem. Phys. 2014 , 16 , 22167-22178]. The CBS limit for the total interaction energy was estimated to be -9.64 kcal/mol.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Improved hydrogeophysical characterization and monitoring through parallel modeling and inversion of time-domain resistivity andinduced-polarization data

Electrical geophysical methods have found wide use in the growing discipline of hydrogeophysics, both for characterizing the electrical properties of the subsurface, and for monitoring subsurface processes in terms of the spatiotemporal changes in subsurface conductivity, chargeability, and source currents they govern. Current multichannel and multielectrode data collections systems are able to collect large amounts of data in relatively short periods of time. However, practitioners are often unable to fully utilize these large data sets and the information they contain due to the processing limitations of standard desktop computers. This limitation can be addressed by utilizing the storage and processing capabilities of high-performance parallel computing environments. We present a parallel distributed-memory forward and inverse modeling algorithm for analyzing resistivity and time-domain induced polarization data. The primary components of the parallel computations include distributed computation of the pole solutions in forward mode, distributed storage and computation of the Jacobian matrix in inverse mode, and parallel execution of the inverse equation solver. We demonstrate the corresponding parallel code for three efforts: (1) resistivity characterization of the Hanford 300 Area Integrated Field Research Challenge site in Hanford, WA; (2) resistivity characterization of a volcanic island in the southern Tyrrhenian Sea in Italy; and (3) resistivity and IP monitoring of biostimulation at a superfund site in Brandywine, MD. Inverse analysis of each of these data sets would be limited (or impossible) in a standard serial computing environment which underscores the need for high-performance computing to fully utilize the potential of electrical geophysical methods in hydrogeophysical applications.

Hanford IFRC↗

Parallel Pascal - An extended Pascal for parallel computers

Parallel Pascal is an extended version of the conventional serial Pascal programming language which includes a convenient syntax for specifying array operations. It is upward compatible with standard Pascal and involves only a small number of carefully chosen new features. Parallel Pascal was developed to reduce the semantic gap between standard Pascal and a large range of highly parallel computers. Two important design goals of Parallel Pascal were efficiency and portability. Portability is particularly difficult to achieve since different parallel computers frequently have very different capabilities.

Reeves, A. P.↗

Directions in parallel programming: HPF, shared virtual memory and object parallelism in pC++

Fortran and C++ are the dominant programming languages used in scientific computation. Consequently, extensions to these languages are the most popular for programming massively parallel computers. We discuss two such approaches to parallel Fortran and one approach to C++. The High Performance Fortran Forum has designed HPF with the intent of supporting data parallelism on Fortran 90 applications. HPF works by asking the user to help the compiler distribute and align the data structures with the distributed memory modules in the system. Fortran-S takes a different approach in which the data distribution is managed by the operating system and the user provides annotations to indicate parallel control regions. In the case of C++, we look at pC++ which is based on a concurrent aggregate parallel model.

Bodin, Francois↗

Data communications in a parallel active messaging interface of a parallel computer

Data communications in a parallel active messaging interface (`PAMI`) of a parallel computer composed of compute nodes that execute a parallel application, each compute node including application processors that execute the parallel application and at least one management processor dedicated to gathering information regarding data communications. The PAMI is composed of data communications endpoints, each endpoint composed of a specification of data communications parameters for a thread of execution on a compute node, including specifications of a client, a context, and a task, the compute nodes and the endpoints coupled for data communications through the PAMI and through data communications resources. Embodiments function by gathering call site statistics describing data communications resulting from execution of data communications instructions and identifying in dependence upon the call cite statistics a data communications algorithm for use in executing a data communications instruction at a call site in the parallel application.

97 MATHEMATICS AND COMPUTING↗

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↗

PFLOTRAN-E4D: A parallel open source PFLOTRAN module for simulating time-lapse electrical resistivity data

Time-lapse electrical resistivity tomography (ERT) is finding increased application for remotely monitoring processes occurring in the near subsurface in three-dimensions (i.e. 4D monitoring). However, there are few codes capable of simulating the evolution of subsurface resistivity and corresponding tomographic measurements arising from a particular process, particularly in parallel and with an open source license. Herein we describe and demonstrate an electrical resistivity tomography module for the PFLOTRAN subsurface simulation code, named PFLOTRAN-E4D. The PFLOTRAN-E4D module operates in parallel using a dedicated set of compute cores in a master-slave configuration. At each time step, the master processes receives subsurface states from PFLOTRAN, converts those states to bulk electrical conductivity, and instructs the slave processes to simulate a tomographic data set. The resulting multi-physics simulation capability enables accurate feasibility studies for ERT imaging, the identification of the ERT signatures that are unique to a given process, and facilitates the joint inversion of ERT data with hydrogeological data for subsurface characterization. PFLOTRAN-E4D is demonstrated herein using a field study of stage-driven groundwater/river water interaction ERT monitoring along the Columbia River, Washington, USA. Results demonstrate the complex nature of changes subsurface electrical conductivity, in both the saturated and unsaturated zones, arising from water table changes and from river water intrusion into the aquifer. The results also demonstrate the sensitivity of surface based ERT measurements to those changes over time. PFLOTRAN-E4D is available with the PFLOTRAN development version with an open-source license at https://bitbucket.org/pflotran/pflotran-dev .

Environmental Molecular Sciences Laboratory↗

A parallel multi-domain solution methodology applied to nonlinear thermal transport problems in nuclear fuel pins

This paper describes an efficient and nonlinearly consistent parallel solution methodology for solving coupled nonlinear thermal transport problems that occur in nuclear reactor applications over hundreds of individual 3D physical subdomains. Efficiency is obtained by leveraging knowledge of the physical domains, the physics on individual domains, and the couplings between them for preconditioning within a Jacobian Free Newton Krylov method. Details of the computational infrastructure that enabled this work, namely the open source Advanced Multi-Physics (AMP) package developed by the authors is described. Details of verification and validation experiments, and parallel performance analysis in weak and strong scaling studies demonstrating the achieved efficiency of the algorithm are presented. Furthermore, numerical experiments demonstrate that the preconditioner developed is independent of the number of fuel subdomains in a fuel rod, which is particularly important when simulating different types of fuel rods. Finally, we demonstrate the power of the coupling methodology by considering problems with couplings between surface and volume physics and coupling of nonlinear thermal transport in fuel rods to an external radiation transport code.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Hybrid parallel programming with MPI and Unified Parallel C.

The Message Passing Interface (MPI) is one of the most widely used programming models for parallel computing. However, the amount of memory available to an MPI process is limited by the amount of local memory within a compute node. Partitioned Global Address Space (PGAS) models such as Unified Parallel C (UPC) are growing in popularity because of their ability to provide a shared global address space that spans the memories of multiple compute nodes. However, taking advantage of UPC can require a large recoding effort for existing parallel applications. In this paper, we explore a new hybrid parallel programming model that combines MPI and UPC. This model allows MPI programmers incremental access to a greater amount of memory, enabling memory-constrained MPI codes to process larger data sets. In addition, the hybrid model offers UPC programmers an opportunity to create static UPC groups that are connected over MPI. As we demonstrate, the use of such groups can significantly improve the scalability of locality-constrained UPC codes. This paper presents a detailed description of the hybrid model and demonstrates its effectiveness in two applications: a random access benchmark and the Barnes-Hut cosmological simulation. Experimental results indicate that the hybrid model can greatly enhance performance; using hybrid UPC groups that span two cluster nodes, RA performance increases by a factor of 1.33 and using groups that span four cluster nodes, Barnes-Hut experiences a twofold speedup at the expense of a 2% increase in code size.

99 GENERAL AND MISCELLANEOUS↗