Vector and Parallel Processing - Vecpar'96: Second International Conference on Vector and Parallel Processing - Systems and Applications, Porto, Portu

ISBN-13: 9783540628286 / Angielski / Miękka / 1997 / 469 str.

Jose Palma; Jose M. L. M. Palma; Jack Dongarra

Opis

This book constitutes a carefully arranged selection of revised full papers chosen from the presentations given at the Second International Conference on Vector and Parallel Processing - Systems and Applications, VECPAR'96, held in Porto, Portugal, in September 1996.
Besides 10 invited papers by internationally leading experts, 17 papers were accepted from the submitted conference papers for inclusion in this documentation following a second round of refereeing. A broad spectrum of topics and applications for which parallelism contributes to progress is covered, among them parallel linear algebra, computational fluid dynamics, data parallelism, implementational issues, optimization, finite element computations, simulation, and visualisation.

Spis treści

High performance computing in power system applications.- Providing access to high performance computing technologies.- Scan-directional architectures.- Markov chain based management of large scale distributed computations of earthen dam leakages.- Irregular data-parallel objects in C++.- Control and data flow analysis for parallel program debugging.- ProHos-1 — A vector processor for the efficient estimation of higher-order moments.- The use of computational kernels in full and sparse linear solvers, efficient code design on high-performance RISC processors.- Parallel implementation of a symmetric eigensolver based on the Yau and Lu method.- Preconditioned conjugate gradient methods for semiconductor device simulation on a CRAY C90 vector processor.- A Parallel implementation of the general Lanczos method on the CRAY T3D.- Parallel and distributed computations in a parameter inverse problem.- Automated optimal design using CFD and high performance computing.- Parallelization of the discrete ordinates method: Two different approaches.- Experiences with advanced CFD algorithms on NEC SX-4.- Parallelization of CFD code using PVM and domain decomposition techniques.- Possibilities of parallel computing in the finite element analysis of industrial forming processes.- Preconditioners for nonsymmetric linear systems in domain decomposition applied to a coupled discretization of Navier-Stokes equations.- Parallel implementation of non recurrent neural networks.- Parallel computing of fragment vector in Steiner triple systems.- Stabilizing large control linear systems on multicomputers.- An interface based on transputers to simulate the dynamic equation of robot manipulators using parallel computing.- Large scale traffic simulations.- Parallel method for automatic shape determination via the evolution of morphological sequences.- Principal component analysis on vector computers.- Functional programming and parallel processing.- A scalable implementation of an interactive increasing realism ray-tracing algorithm.