Parallel Overrelaxation Algorithms for Systems of Linear Equations
نویسنده
چکیده
Parallel versions of the Jacobi and Successive Overrelaxation methods to solve systems of linear equations on a network of transputers are discussed. The Jacobi method is easily parallelizable, by distributing the data in a rowwise fashion among the processors. The SOR is a less obviously parallelizable method due to data-dependency issues, although some computation can still be distributed. An important feature of our implementations is the parallel computation of matrix norms, which allow a significant speed-up over the serial versions of the algorithms. We show a significant improvement in the efficiency of our implementations over other previously published parallel iterative linear equations solvers.
منابع مشابه
GPU computing with Kaczmarz's and other iterative algorithms for linear systems
The graphics processing unit (GPU) is used to solve large linear systems derived from partial differential equations. The differential equations studied are strongly convection-dominated, of various sizes, and common to many fields, including computational fluid dynamics, heat transfer, and structural mechanics. The paper presents comparisons between GPU and CPU implementations of several well-...
متن کاملOn the modified iterative methods for $M$-matrix linear systems
This paper deals with scrutinizing the convergence properties of iterative methods to solve linear system of equations. Recently, several types of the preconditioners have been applied for ameliorating the rate of convergence of the Accelerated Overrelaxation (AOR) method. In this paper, we study the applicability of a general class of the preconditioned iterative methods under certain conditio...
متن کاملExplicit group AOR method for solving elliptic partial differential equations
This paper presents a new 4-points Explicit Group Unsymmetric Successive Overrelaxation (USSOR) iterative method to approximate the solution of the linear systems derived from the discretisation of self-adjoint elliptic partial equations. Several studies have been carried out by many researchers on the USSOR iterative method, for example, the analysis of its convergence [1], an upper bound for ...
متن کاملStatic Task Allocation in Distributed Systems Using Parallel Genetic Algorithm
Over the past two decades, PC speeds have increased from a few instructions per second to several million instructions per second. The tremendous speed of today's networks as well as the increasing need for high-performance systems has made researchers interested in parallel and distributed computing. The rapid growth of distributed systems has led to a variety of problems. Task allocation is a...
متن کاملAccelerated Overrelaxation Method
This paper describes a method for the numerical solution of linear systems of equations. The method is a two-parameter generalization of the Successive Overrelaxation (SOR) method such that when the two parameters involved are equal it coincides with the SOR method. Finally, a numerical example is given to show the superiority of the new method.
متن کامل