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 127 records · Page 7

Parallel and Distributed System Simulation

This exploratory study initiated our research into the software infrastructure necessary to support the modeling and simulation techniques that are most appropriate for the Information Power Grid. Such computational power grids will use high-performance networking to connect hardware, software, instruments, databases, and people into a seamless web that supports a new generation of computation-rich problem solving environments for scientists and engineers. In this context we looked at evaluating the NetSolve software environment for network computing that leverages the potential of such systems while addressing their complexities. NetSolve's main purpose is to enable the creation of complex applications that harness the immense power of the grid, yet are simple to use and easy to deploy. NetSolve uses a modular, client-agent-server architecture to create a system that is very easy to use. Moreover, it is designed to be highly composable in that it readily permits new resources to be added by anyone willing to do so. In these respects NetSolve is to the Grid what the World Wide Web is to the Internet. But like the Web, the design that makes these wonderful features possible can also impose significant limitations on the performance and robustness of a NetSolve system. This project explored the design innovations that push the performance and robustness of the NetSolve paradigm as far as possible without sacrificing the Web-like ease of use and composability that make it so powerful.

Dongarra, Jack↗

Task Assignment Heuristics for Parallel and Distributed CFD Applications

This paper proposes a task graph (TG) model to represent a single discrete step of multi-block overset grid computational fluid dynamics (CFD) applications. The TG model is then used to not only balance the computational workload across the overset grids but also to reduce inter-grid communication costs. We have developed a set of task assignment heuristics based on the constraints inherent in this class of CFD problems. Two basic assignments, the smallest task first (STF) and the largest task first (LTF), are first presented. They are then systematically costs. To predict the performance of the proposed task assignment heuristics, extensive performance evaluations are conducted on a synthetic TG with tasks defined in terms of the number of grid points in predetermined overlapping grids. A TG derived from a realistic problem with eight million grid points is also used as a test case.

Lopez-Benitez, Noe↗

A Portable 3D FFT Package for Distributed-Memory Parallel Architectures

A parallel algorithm for 3D FFTs is implemented as a series of local 1D FFTs combined with data transposes. This allows the use of vendor supplied (often fully optimized) sequential 1D FFTs. The FFTs are carried out in-place by using an in-place data transpose across the processors.

fast Fourrier transform FFT parallel processing↗

Parallel volume ray-casting for unstructured-grid data on distributed-memory architectures

As computing technology continues to advance, computational modeling of scientific and engineering problems produces data of increasing complexity: large in size and unstructured in shape. Volume visualization of such data is a challenging problem. This paper proposes a distributed parallel solution that makes ray-casting volume rendering of unstructured-grid data practical. Both the data and the rendering process are distributed among processors. At each processor, ray-casting of local data is performed independent of the other processors. The global image composing processes, which require inter-processor communication, are overlapped with the local ray-casting processes to achieve maximum parallel efficiency. This algorithm differs from previous ones in four ways: it is completely distributed, less view-dependent, reasonably scalable, and flexible. Without using dynamic load balancing, test results on the Intel Paragon using from two to 128 processors show, on average, about 60% parallel efficiency.

Ma, Kwan-Liu↗

A FAST POTENTIAL AND SELF-GRAVITY SOLVER FOR NONAXISYMMETRIC DISKS

Disk self-gravity could play an important role in the dynamic evolution of interaction between disks and embedded protoplanets. We have developed a fast and accurate solver to calculate the disk potential and disk self-gravity forces for disk systems on a uniform polar grid. Our method closely follows the method given by Chan et al., in which a fast Fourier transform in the azimuthal direction is performed and a direct integral approach in the frequency domain in the radial direction is implemented on a uniform polar grid. This method can be very effective for disks with vertical structures that depend only on the disk radius, achieving the same computational efficiency as for zero-thickness disks. We describe how to parallelize the solver efficiently on distributed parallel computers. We propose a mode-cutoff procedure to reduce the parallel communication cost and achieve nearly linear scalability for a large number of processors. For comparison, we have also developed a particle-based fast tree code to calculate the self-gravity of the disk system with a vertical structure. The numerical results show that our direct integral method is at least two orders of magnitude faster than our optimized tree-code approach.

79 ASTRONOMY AND ASTROPHYSICS↗

Unorthodox parallelization for Bayesian quantum state estimation

Quantum state tomography (QST) allows for the reconstruction of quantum states through measurements and some inference technique under the assumption of repeated state preparations. Bayesian inference provides a promising platform to achieve both efficient QST and accurate uncertainty quantification, yet is generally plagued by the computational limitations associated with long Markov chains. In this work, we present a novel Bayesian QST approach that leverages modern distributed parallel computer architectures to efficiently sample a D-dimensional Hilbert space. Using a parallelized preconditioned Crank–Nicholson Metropolis–Hastings algorithm, we demonstrate our approach on simulated data and experimental results from IBM Quantum systems up to four qubits, showing significant speedups through parallelization. Although highly unorthodox in pooling independent Markov chains, our method proves remarkably practical, with validation ex post facto via diagnostics like the intrachain autocorrelation time. We conclude by discussing scalability to higher-dimensional systems, offering a path toward efficient and accurate Bayesian characterization of large quantum systems.

Bayesian inference↗

Measurements of parallel electron velocity distributions using whistler wave absorption

We describe a diagnostic to measure the parallel electron velocity distribution in a magnetized plasma that is overdense ({omega}{sub pe} > {omega}{sub ce}). This technique utilizes resonant absorption of whistler waves by electrons with velocities parallel to a background magnetic field. The whistler waves were launched and received by a pair of dipole antennas immersed in a cylindrical discharge plasma at two positions along an axial background magnetic field. The whistler wave frequency was swept from somewhat below and up to the electron cyclotron frequency {omega}{sub ce}. As the frequency was swept, the wave was resonantly absorbed by the part of the electron phase space density which was Doppler shifted into resonance according to the relation {omega}-k{sub ||v||} = {omega}{sub ce}. The measured absorption is directly related to the reduced parallel electron distribution function integrated along the wave trajectory. The background theory and initial results from this diagnostic are presented here. Though this diagnostic is best suited to detect tail populations of the parallel electron distribution function, these first results show that this diagnostic is also rather successful in measuring the bulk plasma density and temperature both during the plasma discharge and into the afterglow.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

A Gain Scheduling Wide-Area Damping Controller for the Efficient Integration of Photovoltaic Plant

With photovoltaic generation increasing, the wide-area oscillation has a vital influence on the long-distance power transmission of photovoltaic power. For an efficient integration of photovoltaic generation, a gain scheduling wide-area damping controller (WADC) is designed to guarantee system damping performance under a wide-area signal delay and a changing operating condition, which could be aggravated by the fluctuant and uncertain photovoltaic generation. A linear parameter-varying (LPV) model is proposed to represent all possible operating conditions within varying ranges of the photovoltaic generation and the load. Then, a tensor product (TP) model transformation is applied to transform the LPV model into a TP model. Based on the TP model, a parallel distributed compensation controller (PDC) framework is adopted to design the gain scheduling WADC, which is solved by linear matrix inequalities (LMIs) considering the accommodation to the random interval delay. In addition, a staircase basis function method is proposed to reduce the conservatism of LMIs within the PDC framework. The gain of the proposed WADC will be adjusted according to the operating condition acknowledged through the real-time scheduling and control center, once a wide-area oscillation is detected. Lastly, the performance of the proposed WADC is verified by a 2-area 4-machine system and a 68-bus 16-machine system.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗

Parallel Implicit Algorithms for CFD

The main goal of this project was efficient distributed parallel and workstation cluster implementations of Newton-Krylov-Schwarz (NKS) solvers for implicit Computational Fluid Dynamics (CFD.) "Newton" refers to a quadratically convergent nonlinear iteration using gradient information based on the true residual, "Krylov" to an inner linear iteration that accesses the Jacobian matrix only through highly parallelizable sparse matrix-vector products, and "Schwarz" to a domain decomposition form of preconditioning the inner Krylov iterations with primarily neighbor-only exchange of data between the processors. Prior experience has established that Newton-Krylov methods are competitive solvers in the CFD context and that Krylov-Schwarz methods port well to distributed memory computers. The combination of the techniques into Newton-Krylov-Schwarz was implemented on 2D and 3D unstructured Euler codes on the parallel testbeds that used to be at LaRC and on several other parallel computers operated by other agencies or made available by the vendors. Early implementations were made directly in Massively Parallel Integration (MPI) with parallel solvers we adapted from legacy NASA codes and enhanced for full NKS functionality. Later implementations were made in the framework of the PETSC library from Argonne National Laboratory, which now includes pseudo-transient continuation Newton-Krylov-Schwarz solver capability (as a result of demands we made upon PETSC during our early porting experiences). A secondary project pursued with funding from this contract was parallel implicit solvers in acoustics, specifically in the Helmholtz formulation. A 2D acoustic inverse problem has been solved in parallel within the PETSC framework.

Keyes, David E.↗

Camellia: A Rapid Development Framework for Finite Element Solvers

We report the discontinuous Petrov–Galerkin (DPG) methodology of Demkowicz and Gopalakrishnan guarantees the optimality of the finite element solution in a user-controllable energy norm, and provides several features supporting adaptive schemes. The approach provides stability automatically; there is no need for carefully derived numerical fluxes (as in DG schemes) or for mesh-dependent stabilization terms (as in stabilized methods). In this paper, we focus on features of Camellia that facilitate implementation of new DPG formulations; chief among these is a rich set of features in support of symbolic manipulation, which allow, e.g., bilinear formulations in the code to appear much as they would on paper. Many of these features are general in the sense that they can also be used in the implementation of other finite element formulations. In fact, because DPG’s requirements are essentially a superset of those of other finite element methods, Camellia provides built-in support for most common methods. We believe, however, that the combination of an essentially “hands-free” finite element methodology as found in DPG with the rapid development features of Camellia are particularly winsome, so we focus on use cases in this class. In addition to the symbolic manipulation features mentioned above, Camellia offers support for one-irregular adaptive meshes in 1D, 2D, 3D, and space-time. It provides a geometric multigrid preconditioner particularly suited for DPG problems, and supports distributed parallel execution using MPI. For its load balancing and distributed data structures, Camellia relies on packages from the Trilinos project, which simplifies interfacing with other computational science packages. Camellia also allows loading of standard mesh formats through an interface with the MOAB package. Camellia includes support for static condensation to eliminate element-interior degrees of freedom locally, usually resulting in substantial reduction of the cost of the global problem. Finally, we include a discussion of the variational formulations built into Camellia, with references to those formulations in the literature, as well as an MPI performance study.

97 MATHEMATICS AND COMPUTING↗

A distributed-memory approximation algorithm for maximum weight perfect bipartite matching

We design and implement an efficient parallel approximation algorithm for the problem of maximum weight perfect matching in bipartite graphs, i.e. the problem of finding a set of non-adjacent edges that covers all vertices and has maximum weight. This problem differs from the maximum weight matching problem, for which scalable approximation algorithms are known. It is primarily motivated by finding good pivots in scalable sparse direct solvers before factorization where sequential implementations of maximum weight perfect matching algorithms, such as those available in MC64, are widely used due to the lack of scalable alternatives. To overcome this limitation, we propose a fully parallel distributed memory algorithm that first generates a perfect matching and then searches for weightaugmenting cycles of length four in parallel and iteratively augments the matching with a vertex disjoint set of such cycles. For most practical problems the weights of the perfect matchings generated by our algorithm are very close to the optimum. An efficient implementation of the algorithm scales up to 256 nodes (17,408 cores) on a Cray XC40 supercomputer and can solve instances that are too large to be handled by a single node using the sequential algorithm.

Azad, Ariful↗

SIERRA Toolkit v. 1.0

The SIERRA Toolkit is a collection of libraries to facilitate the development of parallel engineering analysis applications. These libraries supply basic core services that an engineering analysis application may need such as a parallel distributed and dynamic mesh database (for unstructured meshes), mechanics algorithm support (parallel infrastructure only), interfaces to parallel solvers, parallel mesh and data I/O, and various utilities (timers, diagnostic tools, etc.).The toolkit is intended to reduce the effort required to develop an engineering analysis application by removing the need to develop core capabilities that most every application would require.

Coffey, Todd↗