An Efficient Parallel Multigrid Solver for 3-D Convection-dominated Problems PDF Download
Are you looking for read ebook online? Search for your book and save it on your Kindle device, PC, phones or tablets. Download An Efficient Parallel Multigrid Solver for 3-D Convection-dominated Problems PDF full book. Access full book title An Efficient Parallel Multigrid Solver for 3-D Convection-dominated Problems by Ignacio M. Llorente. Download full books in PDF and EPUB format.
Author: Ignacio M. Llorente
Publisher:
ISBN:
Category : Boundary value problems
Languages : en
Pages : 36
Get Book Here
Book Description
Multigrid algorithms are known to be highly efficient in solving systems of elliptic equations. However, standard multi grid algorithms fail to achieve optimal grid-independent convergence rates in solving non-elliptic problems. In many practical cases, the non-elliptic part of a problem is represented by the convection operator. Downstream marching, when it is viable, is the simplest and most efficient way to solve this operator. However, in a parallel setting, the sequential nature of marching degrades the efficiency of the algorithm. The aim of this report is to present, evaluate and analyze an alternative highly parallel multi grid method for 3-D convection-dominated problems. This method employs semi coarsening, a four-color plane-implicit smoother, and discretization rules allowing the same cross-characteristic interactions on all the grids involved to be maintained. The resulting multigrid solver exhibits a fast grid-independent convergence rate for solving the convection-diffusion operator on cell-centered grids with stretching. The load imbalance below the critical level is the main source of inefficiency in its parallel implementation. A hybrid smoother that degrades the convergence properties of the method but improves its granularity has been found to be the best choice in a parallel setting. The numerical and parallel properties of the multi grid algorithm with the four-color and hybrid smoothers are studied on SGI Origin 2000 and Cray T3E systems.
Author: Ignacio M. Llorente
Publisher:
ISBN:
Category : Boundary value problems
Languages : en
Pages : 36
Get Book Here
Book Description
Multigrid algorithms are known to be highly efficient in solving systems of elliptic equations. However, standard multi grid algorithms fail to achieve optimal grid-independent convergence rates in solving non-elliptic problems. In many practical cases, the non-elliptic part of a problem is represented by the convection operator. Downstream marching, when it is viable, is the simplest and most efficient way to solve this operator. However, in a parallel setting, the sequential nature of marching degrades the efficiency of the algorithm. The aim of this report is to present, evaluate and analyze an alternative highly parallel multi grid method for 3-D convection-dominated problems. This method employs semi coarsening, a four-color plane-implicit smoother, and discretization rules allowing the same cross-characteristic interactions on all the grids involved to be maintained. The resulting multigrid solver exhibits a fast grid-independent convergence rate for solving the convection-diffusion operator on cell-centered grids with stretching. The load imbalance below the critical level is the main source of inefficiency in its parallel implementation. A hybrid smoother that degrades the convergence properties of the method but improves its granularity has been found to be the best choice in a parallel setting. The numerical and parallel properties of the multi grid algorithm with the four-color and hybrid smoothers are studied on SGI Origin 2000 and Cray T3E systems.
Author: Manuel Prieto
Publisher:
ISBN:
Category : Multigrid methods (Numerical analysis)
Languages : en
Pages : 26
Get Book Here
Book Description
Abstract: "This paper presents an efficient parallel multigrid solver for speeding up the computation of a 3-D model that treats the flow of a viscous fluid over a flat plate. The main interest of this simulation lies in exhibiting some basic difficulties that prevent optimal multigrid efficiencies from being achieved. As the computing platform, we have used Coral, a Beowulf-class system based on Intel Pentium processors and equipped with GigaNet cLAN and switched Fast Ethernet networks. Our study not only examines the scalability of the solver but also includes a performance evaluation of Coral where the investigated solver has been used to compare several of its design choices, namely, the interconnection network (GigaNet versus switched Fast-Ethernet) and the node configuration (dual nodes versus single nodes). As a reference, the performance results have been compared with those obtained with the NAS-MG benchmark."
Author: Anton Schüller
Publisher: Springer Science & Business Media
ISBN: 3322865762
Category : Technology & Engineering
Languages : en
Pages : 232
Get Book Here
Book Description
This book contains the main results of the German project POPINDA. It surveys the state of the art of industrial aerodynamic design simulations on parallel systems. POPINDA is an acronym for Portable Parallelization of Industrial Aerodynamic Applications. This project started in late 1993. The research and development work invested in POPINDA corresponds to about 12 scientists working full-time for the three and a half years of the project. POPINDA was funded by the German Federal Ministry for Education, Science, Research and Technology (BMBF). The central goals of POPINDA were to unify and parallelize the block-structured aerodynamic flow codes of the German aircraft industry and to develop new algorithmic approaches to improve the efficiency and robustness of these programs. The philosophy behind these goals is that challenging and important numerical appli cations such as the prediction of the 3D viscous flow around full aircraft in aerodynamic design can only be carried out successfully if the benefits of modern fast numerical solvers and parallel high performance computers are combined. This combination is a "conditio sine qua non" if more complex applications such as aerodynamic design optimization or fluid structure interaction problems have to be solved. When being solved in a standard industrial aerodynamic design process, such more complex applications even require a substantial further reduction of computing times. Parallel and vector computers on the one side and innovative numerical algorithms such as multigrid on the other have enabled impressive improvements in scientific computing in the last 15 years.
Author: C.B. Jenssen
Publisher: Gulf Professional Publishing
ISBN: 0080538401
Category : Science
Languages : en
Pages : 601
Get Book Here
Book Description
Parallel CFD 2000, the Twelfth in an International series of meetings featuring computational fluid dynamics research on parallel computers, was held May 22-25, 2000 in Trondheim, Norway.Following the trend of the past conferences, areas such as numerical schemes and algorithms, tools and environments, load balancing, as well as interdisciplinary topics and various kinds of industrial applications were all well represented in the work presented. In addition, for the first time in the Parallel CFD conference series, the organizing committee chose to draw special attention to certain subject areas by organizing a number of special sessions.We feel the emphasis of the papers presented at the conference reflect the direction of the research within parallel CFD at the beginning of the new millennium. It seems to be a clear tendency towards increased industrial exploitation of parallel CFD. Several presentations also demonstrated how new insight is being achieved from complex simulations, and how powerful parallel computers now make it possible to use CFD within a broader interdisciplinary setting.Obviously, successful application of parallel CFD still rests on the underlying fundamental principles. Therefore, numerical algorithms, development tools, and parallelization techniques are still as important as when parallel CFD was in is infancy. Furthermore, the novel concepts of affordable parallel computing as well as metacomputing show that exciting developments are still taking place.As is often pointed out however, the real power of parallel CFD comes from the combination of all the disciplines involved: Physics, mathematics, and computer science. This is probably one of the principal reasons for the continued popularity of the Parallel CFD Conferences series, as well as the inspiration behind much of the excellent work carried out on the subject. We hope that the papers in this book, both on an individual basis and as a whole, will contribute to that inspiration. Further details of Parallel CFD'99, as well as other conferences in this series, are available at http://www.parcfd.org
Author: Ruben S. Montero
Publisher:
ISBN:
Category : Multigrid methods (Numerical analysis)
Languages : en
Pages : 24
Get Book Here
Book Description
This paper presents a full multigrid solver for the simulation of flow over a yawed at plate. The two problems associated with this simulation; boundary layers and entering flows with non-aligned characteristics, have been successfully overcome through the combination of a plane-implicit solver and semicoarsening. In fact, this multigrid algorithm exhibits a textbook multigrid convergence rate, i.e., the solution of the discrete system of equations is obtained in a fixed amount of computational work, independently of the grid size, grid stretching factor and non-alignment parameter. Also, a parallel variant of the smoother based on a four-color ordering of planes is investigated.
Author: Glyn Owen Rees
Publisher:
ISBN:
Category :
Languages : en
Pages :
Get Book Here
Book Description
Author: Ute Gärtel
Publisher:
ISBN:
Category : Multiprocessors
Languages : en
Pages : 14
Get Book Here
Book Description
Abstract: "A parallel multigrid program for solving 3D elliptic problems with arbitrary anisotropies has been implemented. For efficient error smoothing, parallel line relaxation is performed by use of a reduction method and plane relaxation is carried out by using suitable parallel 2D multigrid methods. For all process creation and communication tasks routines of the SUPRENUM communications library are called. This yields full user program portability among all machines for which the library has been implemented. Numerical results, especially concerning multiprocessor efficiencies, are presented and discussed."
Author:
Publisher:
ISBN:
Category :
Languages : en
Pages : 74
Get Book Here
Book Description
Author:
Publisher:
ISBN:
Category : Mathematical statistics
Languages : en
Pages : 838
Get Book Here
Book Description
Author:
Publisher:
ISBN:
Category :
Languages : en
Pages : 162
Get Book Here
Book Description
