Parallel Sub-structuring Methods for Solving Sparse Linear Systems on a Cluster of GPUs
A. Cheik Ahamed, und F. Magoulès. High Performance Computing and Communications, 2014 IEEE 6th Intl Symp on Cyberspace Safety and Security, 2014 IEEE 11th Intl Conf on Embedded Software and Syst (HPCC,CSS,ICESS), 2014 IEEE Intl Conf on, Seite 121-128. (August 2014)
DOI: 10.1109/HPCC.2014.24
Zusammenfassung
The main objective of this work consists in analyzing sub-structuring method for the parallel solution of sparse linear systems with matrices arising from the discretization of partial differential equations such as finite element, finite volume and finite difference. With the success encountered by the general-purpose processing on graphics processing units (GPGPU), we develop an hybrid multi GPUs and CPUs sub-structuring algorithm. GPU computing, with CUDA, is used to accelerate the operations performed on each processor. Numerical experiments have been performed on a set of matrices arising from engineering problems. We compare C+MPI implementation on classical CPU cluster with C+MPI+CUDA on a cluster of GPU. The performance comparison shows a speed-up for the sub-structuring method up to 19 times in double precision by using CUDA.
High Performance Computing and Communications, 2014 IEEE 6th Intl Symp on Cyberspace Safety and Security, 2014 IEEE 11th Intl Conf on Embedded Software and Syst (HPCC,CSS,ICESS), 2014 IEEE Intl Conf on
%0 Conference Paper
%1 cheikahamed2014parallel
%A Cheik Ahamed, Abal-Kassim
%A Magoulès, Frédéric
%B High Performance Computing and Communications, 2014 IEEE 6th Intl Symp on Cyberspace Safety and Security, 2014 IEEE 11th Intl Conf on Embedded Software and Syst (HPCC,CSS,ICESS), 2014 IEEE Intl Conf on
%D 2014
%K CUDA Computing Conjugate Finite GPU Gradient Graphics Linear Parallel Processing Sub-structuring Unit algebra and computing dblp distributed element method myown
%P 121-128
%R 10.1109/HPCC.2014.24
%T Parallel Sub-structuring Methods for Solving Sparse Linear Systems on a Cluster of GPUs
%U http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=7056727
%X The main objective of this work consists in analyzing sub-structuring method for the parallel solution of sparse linear systems with matrices arising from the discretization of partial differential equations such as finite element, finite volume and finite difference. With the success encountered by the general-purpose processing on graphics processing units (GPGPU), we develop an hybrid multi GPUs and CPUs sub-structuring algorithm. GPU computing, with CUDA, is used to accelerate the operations performed on each processor. Numerical experiments have been performed on a set of matrices arising from engineering problems. We compare C+MPI implementation on classical CPU cluster with C+MPI+CUDA on a cluster of GPU. The performance comparison shows a speed-up for the sub-structuring method up to 19 times in double precision by using CUDA.
@inproceedings{cheikahamed2014parallel,
abstract = {The main objective of this work consists in analyzing sub-structuring method for the parallel solution of sparse linear systems with matrices arising from the discretization of partial differential equations such as finite element, finite volume and finite difference. With the success encountered by the general-purpose processing on graphics processing units (GPGPU), we develop an hybrid multi GPUs and CPUs sub-structuring algorithm. GPU computing, with CUDA, is used to accelerate the operations performed on each processor. Numerical experiments have been performed on a set of matrices arising from engineering problems. We compare C+MPI implementation on classical CPU cluster with C+MPI+CUDA on a cluster of GPU. The performance comparison shows a speed-up for the sub-structuring method up to 19 times in double precision by using CUDA.},
added-at = {2015-03-18T14:57:54.000+0100},
author = {Cheik Ahamed, Abal-Kassim and Magoulès, Frédéric},
biburl = {https://www.bibsonomy.org/bibtex/2f565e355262da41967144158e54bfd3a/akcheik},
booktitle = {High Performance Computing and Communications, 2014 IEEE 6th Intl Symp on Cyberspace Safety and Security, 2014 IEEE 11th Intl Conf on Embedded Software and Syst (HPCC,CSS,ICESS), 2014 IEEE Intl Conf on},
doi = {10.1109/HPCC.2014.24},
interhash = {7128496bb215ef53d3d7bb8c843b1fec},
intrahash = {f565e355262da41967144158e54bfd3a},
keywords = {CUDA Computing Conjugate Finite GPU Gradient Graphics Linear Parallel Processing Sub-structuring Unit algebra and computing dblp distributed element method myown},
month = aug,
pages = {121-128},
timestamp = {2015-03-20T22:19:19.000+0100},
title = {Parallel Sub-structuring Methods for Solving Sparse Linear Systems on a Cluster of GPUs},
url = {http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=7056727},
year = 2014
}