Benchmarking Performance of Parallel Computers Using a 2d Elliptic Solver
نویسندگان
چکیده
It was recently shown that block-circulant preconditioners applied to a conjugate gradient method used to solve structured sparse linear systems arising from 2D elliptic problems have very good numerical properties and a potential for good parallel efficiency. The aim of the presentation is to summarize and compare their parallel performance across a number of modern parallel computers: SGI Power Challenge 10000, SGI/Cray Origin 2000, HP-Convex SPP-2000, IBM SP-2, Cray T3E, a network of SUN workstations and a Beowulf cluster of PC’s.
منابع مشابه
Fast Elliptic Solver for Incompressible Navier Stokes Flow and Heat Transfer Problems on the Grid
The objective of this work is to present a fast parallel elliptic solver that improves efficiently the run of incompressible Navier-Stokes Flow code or Heat Transfer code on a grid of parallel computers. We focus on the design of the elliptic solver because the pressure solver in an incompressible Navier-Stokes code , is the one that is most time consuming, and demanding on communications betwe...
متن کاملParallel Performance of a 3D Elliptic Solver
It was recently shown that block-circulant preconditioners applied to a conjugate gradient method used to solve structured sparse linear systems arising from 2D or 3D elliptic problems have good numerical properties and a potential for high parallel efficiency. In this note parallel performance of a circulant block-factorization based preconditioner applied to a 3D model problem is investigated...
متن کاملBenchmarking Performance Analysis of Parallel Solver for 3D Elasticity Problems
In this paper we consider numerical solution of 3D linear elasticity equations described by a coupled system of second order elliptic partial differential equations. This system is discretized by trilinear parallelepipedal finite elements. Preconditioned Conjugate Gradient iterative method is used for solving large-scale linear algebraic systems arising after the Finite Element Method (FEM) dis...
متن کاملComparative Performance of Frontal (Direct) and PCG (Iterative) Solver Based Parallel Computations of Finite Element Analysis
PARAM-10000 is latest of the series of High Performance Parallel Super Computers developed by India. It employs a distributed memory and message passing architectures and is built as a cluster of SMP workstations using 4-way SMP workstations from Sun Microsystems. The work discusses about Parallel Finite element Analysis Programs developed with Direct Frontal Solver and Iterative PCG (Pre-condi...
متن کاملEquilibrium condition nonlinear modeling of a cracked concrete beam using a 2D Galerkin finite volume solver
A constitutive model based on two–dimensional unstructured Galerkin finite volume method (GFVM) is introduced and applied for analyzing nonlinear behavior of cracked concrete structures in equilibrium condition. The developed iterative solver treats concrete as an orthotropic nonlinear material and considers the softening and hardening behavior of concrete under compression and tension by using...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2013