High Performance Computing for Computational Science - VECPAR 2002

• Tipo: Livro
• Data: 2003

Autores

José M. L. M. Palma (Editor) FEUP; Jack Dongarra (Editor) FEUP; Vicente Hernández (Editor) FEUP; António Augusto de Sousa (Editor) FEUP

Editora

Springer

Cham

Indexação

ISI Proceedings

Classificação Científica

FOS: Ciências exactas e naturais > Ciências da computação e da informação

Outras Informações

Abstract (EN): The proceedings contain 49 papers. The special focus in this conference is on Fluids, structures and Data mining. The topics include: Fluid-structure interaction simulations using parallel computers; investigation of turbulent flame kernels using DNS on clusters; A parallel, state-of-the-art, least-squares spectral element solver for incompressible flow problems; edge-based interface elements for solution of three-dimensional geomechanical problems; finite element simulation of seismic elastic two dimensional wave propagation; parallel implementation for probabilistic analysis of 3D discrete cracking in concrete; an a posteriori error estimator for adaptive mesh refinement using parallel in-element particle tracking methods; high performance data mining; data mining for data classification based on the KNN-fuzzy method supported by genetic algorithm; a major challenge for computational chemistry; high performance computing in electron microscope tomography of complex biological structures; visualization of RNA pseudoknot structures; the cactus framework and toolkit; performance of message-passing MATLAB toolboxes; evaluating the performance of space plasma simulations using FPGA’s; remote parallel model reduction of linear time-invariant systems made easy; an approach to teaching computer arithmetic; fast sparse matrix-vector multiplication for TeraFlop/s computers; performance evaluation of parallel gram-Schmidt re-orthogonalization methods; toward memory-efficient linear solvers; a parallel Newton-GMRES algorithm for solving large scale nonlinear systems; preconditioning for an iterative elliptic solver on a vector processor; a large scale electron scattering simulation dominated by the multiplication of dynamically changing matrices; a parallel implementation of the Atkinson algorithm for solving a fredholm equation.

Idioma: Inglês

Tipo (Avaliação Docente): Científica

Nº de páginas: 734

ISBN: 3-540-00852-7