Search results for "Computational Science"
showing 10 items of 124 documents
Building blocks for odd–even multigrid with applications to reduced systems
2001
Abstract Building blocks yielding an efficient implementation of the odd–even multigrid method for the Poisson problem in the reference domain (0,1) d , d=2,3, are described. Modifications needed to transform these techniques to solve reduced linear systems representing boundary value problems in arbitrary domains are given. A new way to define enriched coarser subspaces in the multilevel realization is proposed. Numerical examples demonstrating the efficiency of developed multigrid methods are included.
Real-time Sound Source Localization on Graphics Processing Units
2013
Abstract Sound source localization is an important topic in microphone array signal processing applications, such as camera steering systems, human-machine interaction or surveillance systems. The Steered Response Power with Phase Transform (SRP- PHAT) algorithm is one of the most well-known approaches for sound source localization due to its good performance in noisy and reverberant environments. The algorithm analyzes the sound power captured by a microphone array on a grid of spatial points in a given room. While localization accuracy can be improved by using a high resolution spatial grid and a high number of microphones, performing the localization task in these circumstances requires …
One-dimensional hydrodynamic modeling of coronal plasmas on transputer arrays
1990
Abstract We describe a concurrent implementation of the Palermo-Harvard hydrodynamic code on cost-effective and modularity expandable transputer arrays. We have tested the effectiveness of our approach by simulating an already well-studied compact solar-flare model on different transputer configurations and compared their performances with those of other machines. We have found that the speed of the concurrent program on a 16-T800 transputers array is ~1/9 of that of the equivalent code optimized for a CRAY X-MP/48. This work clearly shows that transputer-based arrays provide locally available high computing-power tools to extend the investigation of compact solar flares and similar astroph…
Millimeter-Scale and Billion-Atom Reactive Force Field Simulation on Sunway Taihulight
2020
Large-scale molecular dynamics (MD) simulations on supercomputers play an increasingly important role in many research areas. With the capability of simulating charge equilibration (QEq), bonds and so on, Reactive force field (ReaxFF) enables the precise simulation of chemical reactions. Compared to the first principle molecular dynamics (FPMD), ReaxFF has far lower requirements on computational resources so that it can achieve higher efficiencies for large-scale simulations. In this article, we present our efforts on scaling ReaxFF on the Sunway TaihuLight Supercomputer (TaihuLight). We have carefully redesigned the force analysis and neighbor list building steps. By applying fine-grained …
VIBPACK: A package to treat multidimensional electron-vibrational molecular problems with application to magnetic and optical properties
2018
We present a FORTRAN code based on a new powerful and efficient computational approach to solve multidimensional dynamic Jahn-Teller and pseudo Jahn-Teller problems. This symmetry-assisted approach constituting a theoretical core of the program is based on the full exploration of the point symmetry of the electronic and vibrational states. We also report some selected examples of increasing complexity aimed to display the theoretical background as well as the advantages and capabilities of the program to evaluate of the energy pattern, magnetic and optical properties of large multimode vibronic systems. © 2018 Wiley Periodicals, Inc.
Imaginary time propagation code for large-scale two-dimensional eigenvalue problems in magnetic fields
2013
We present a code for solving the single-particle, time-independent Schr\"odinger equation in two dimensions. Our program utilizes the imaginary time propagation (ITP) algorithm, and it includes the most recent developments in the ITP method: the arbitrary order operator factorization and the exact inclusion of a (possibly very strong) magnetic field. Our program is able to solve thousands of eigenstates of a two-dimensional quantum system in reasonable time with commonly available hardware. The main motivation behind our work is to allow the study of highly excited states and energy spectra of two-dimensional quantum dots and billiard systems with a single versatile code, e.g., in quantum …
GPU accelerated Monte Carlo simulation of the 2D and 3D Ising model
2009
The compute unified device architecture (CUDA) is a programming approach for performing scientific calculations on a graphics processing unit (GPU) as a data-parallel computing device. The programming interface allows to implement algorithms using extensions to standard C language. With continuously increased number of cores in combination with a high memory bandwidth, a recent GPU offers incredible resources for general purpose computing. First, we apply this new technology to Monte Carlo simulations of the two dimensional ferromagnetic square lattice Ising model. By implementing a variant of the checkerboard algorithm, results are obtained up to 60 times faster on the GPU than on a curren…
Fourth Workshop on using Emerging Parallel Architectures
2012
AbstractThe Fourth Workshop on Using Emerging Parallel Architectures (WEPA), held in conjunction with ICCS 2012, provides a forum for exploring the capabilities of emerging parallel architectures such as GPUs, FPGAs, Cell B.E., Intel M.I.C. and multicores to accelerate computational science applications.
Bounds on transient phase plastic deformations in optimal design of steel frames subjected to cyclic loads
2009
A minimum volume multicriterion design of elastic perfectly plastic steel frames subjected to a combination of quasi-static fixed and cyclic loads, bounding the transient phase plastic deformations, is proposed. The problem is formulated according to a plastic shakedown criterion, so that incremental and instantaneous collapse are certainly prevented when the frame is subjected to very strongly amplified cyclic loads. The further condition that the structure must also behave elastically in serviceability conditions is imposed. Since the steady-state loading history is known it is possible to directly bound the steady-state plastic deformations. By applying a suitable own bounding theorem, i…
VBFNLO: A parton level Monte Carlo for processes with electroweak bosons
2009
VBFNLO is a fully flexible parton level Monte Carlo program for the simulation of vector boson fusion, double and triple vector boson production in hadronic collisions at next-to-leading order in the strong Coupling constant. VBFNLO includes Higgs and vector boson decays with full spin correlations and all off-shell effects. In addition, VBFNLO implements CP-even and CP-odd Higgs boson via gluon fusion, associated with two jets, at the leading-order one-loop level with the full top- and bottom-quark mass dependence in a generic two-Higgs-doublet model. A variety of effects arising from beyond the Standard Model physics are implemented for selected processes. This includes anomalous coupling…