A parallel algebraic multigrid solver on graphics processing units

Authors:
Gundolf Haase;Manfred Liebmann;Craig C. Douglas;Gernot Plank
Affiliations:
Institute for Mathematics and Scientific Computing, University of Graz;Institute for Mathematics and Scientific Computing, University of Graz;Department of Mathematics, University of Wyoming;Computing Laboratory, Oxford University
Venue:
HPCA'09 Proceedings of the Second international conference on High Performance Computing and Applications
Year:
2009

Citing 7
Cited 8

Using MPI (2nd ed.): portable parallel programming with the message-passing interface

Using MPI (2nd ed.): portable parallel programming with the message-passing interface
A multigrid tutorial (2nd ed.)

A multigrid tutorial (2nd ed.)
Parallel Algebraic Multigrid Methods on Distributed Memory Computers

SIAM Journal on Scientific Computing
Tutorial on Elliptic PDE Solvers and Their Parallelization

Tutorial on Elliptic PDE Solvers and Their Parallelization
Sparsity: Optimization Framework for Sparse Matrix Kernels

International Journal of High Performance Computing Applications
Using GPUs to improve multigrid solver performance on a cluster

International Journal of Computational Science and Engineering
Multigrid Methods on GPUs

Multigrid Methods on GPUs

Sparse approximate inverse preconditioners for iterative solvers on GPUs

Proceedings of the 2012 Symposium on High Performance Computing
Efficient AMG on heterogeneous systems

Facing the Multicore-Challenge II
A GPU-Accelerated parallel preconditioner for the solution of the boltzmann transport equation for semiconductors

Facing the Multicore-Challenge II
Parallel smoothers for matrix-based geometric multigrid methods on locally refined meshes using multicore CPUs and GPUs

Facing the Multicore-Challenge II
Parallel geometric-algebraic multigrid on unstructured forests of octrees

SC '12 Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis
Efficient setup of aggregation AMG for CFD on GPUs

PARA'12 Proceedings of the 11th international conference on Applied Parallel and Scientific Computing
Using LAMA for efficient AMG on hybrid clusters

Computer Science - Research and Development
Architecting the finite element method pipeline for the GPU

Journal of Computational and Applied Mathematics

Quantified Score

Hi-index	0.01

Visualization

Abstract

The paper presents a multi-GPU implementation of the preconditioned conjugate gradient algorithm with an algebraic multigrid preconditioner (PCG-AMG) for an elliptic model problem on a 3D unstructured grid. An efficient parallel sparse matrix-vector multiplication scheme underlying the PCG-AMG algorithm is presented for the many-core GPU architecture. A performance comparison of the parallel solver shows that a singe Nvidia Tesla C1060 GPU board delivers the performance of a sixteen node Infiniband cluster and a multi-GPU configuration with eight GPUs is about 100 times faster than a typical server CPU core.