Name
Title | ||
|---|---|---|
Solving 2D Nonlinear Unsteady Convection-Diffusion Equations on Heterogenous Platforms with Multiple GPUs |
Abstract | ||
|---|---|---|
Solving complex convection-diffusion equations is very important to many practical mathematical and physical problems. After the finite difference discretization, most of the time for equations solution is spent on sparse linear equation solvers. In this paper, our goal is to solve 2D Nonlinear Unsteady Convection-Diffusion Equations by accelerating an iterative algorithm named Jacobi-preconditioned QMRCGSTAB on a heterogenous platform, which is composed of a multi-core processor and multiple GPUs. Firstly, a basic implementation and evaluation for adapting the problem to this kind of platform is given. Then, we propose two optimization methods to improve the performance: kernel merging method and matrix boundary data processing. Our experimental evaluation on an AMD Opteron(tm) quad-core processor 2380 linked to an NVIDIA Tesla S1070 platform with four GPUs delivers the peak performance of 33 GFLOPS (double precision), which is a speedup of close to a factor 32 compared to the same problem running on 4 cores of the same CPU. |
Year | DOI | Venue |
|---|---|---|
2009 | 10.1109/ICPADS.2009.76 | ICPADS |
Keywords | Field | DocType |
matrix boundary data processing,sparse linear equation solvers,peak performance,nonlinear unsteady convection-diffusion equations,multi-core processor,nvidia tesla s1070 platform,gpu,heterogenous platform,accelerating iterative algorithm,kernel merging method,heterogenous,accelerate,computer graphics,mathematical analysis,multicore processor,2d nonlinear unsteady convection-diffusion equations,multiple gpu,equations solution,physical problem,jacobi preconditioned qmrcgstab,heterogenous platforms,amd opteron,finite difference discretization,quad-core processor,multiple gpus,experimental evaluation,coprocessors,finite difference methods,nucde solver,quad core processor,pqmrcgstab,iterative methods,finite difference,sparse matrices,multi core processor,linear equations,convection diffusion equation,iterative algorithm,optimization,kernel,merging,data processing | Linear equation,Nonlinear system,Computer science,Iterative method,Parallel computing,Finite difference method,Coprocessor,Multi-core processor,Sparse matrix,Speedup | Conference |
ISSN | ISBN | Citations |
1521-9097 | 978-1-4244-5788-5 | 0 |
PageRank | References | Authors |
0.34 | 6 | 5 |