Block-Iterative Methods for 3D Constant-Coefficient Stencils on GPUs and Multicore CPUs

doi:https://doi.org/10.2172/1134156

Title: Block-Iterative Methods for 3D Constant-Coefficient Stencils on GPUs and Multicore CPUs

Full Record
Other Related Research

Abstract

Block iterative methods are extremely important as smoothers for multigrid methods, as preconditioners for Krylov methods, and as solvers for diagonally dominant linear systems. Developing robust and efficient smoother algorithms suitable for current and evolving GPU and multicore CPU systems is a significant challenge. We address this issue in the case of constant-coefficient stencils arising in the solution of elliptic partial differential equations on structured 3D uniform and adaptively refined block structured grids. Robust, highly parallel implementations of block Jacobi and chaotic block Gauss-Seidel algorithms with exact inversion of the blocks are developed using different parallelization techniques. Experimental results for NVIDIA Fermi/Kepler GPUs and AMD multicore systems are presented.

Authors:

Philip, Bobby ^[1]; Wang, Zhen ^[1]; Berrill, Mark A ^[1]

ORNL

Publication Date:: 2014-06-01

Research Org.:: Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States). Oak Ridge Leadership Computing Facility (OLCF)

Sponsoring Org.:: USDOE Office of Science (SC)

OSTI Identifier:: 1134156

Report Number(s):: ORNL/TM-2014/225
KJ0403000; ERKJR04

DOE Contract Number:: DE-AC05-00OR22725

Resource Type:: Technical Report

Country of Publication:: United States

Language:: English

Citation Formats


                    Philip, Bobby, Wang, Zhen, and Berrill, Mark A. Block-Iterative Methods for 3D Constant-Coefficient Stencils on GPUs and Multicore CPUs.  United States: N. p., 2014. 
        Web.  doi:10.2172/1134156.

Copy to clipboard


                    Philip, Bobby, Wang, Zhen, & Berrill, Mark A. Block-Iterative Methods for 3D Constant-Coefficient Stencils on GPUs and Multicore CPUs.  United States.  https://doi.org/10.2172/1134156

Copy to clipboard


                    Philip, Bobby, Wang, Zhen, and Berrill, Mark A. Sun .  
        "Block-Iterative Methods for 3D Constant-Coefficient Stencils on GPUs and Multicore CPUs".  United States.  https://doi.org/10.2172/1134156.

Copy to clipboard


                    
@article{osti_1134156,

  title        = {Block-Iterative Methods for 3D Constant-Coefficient Stencils on GPUs and Multicore CPUs},

  author       = {Philip, Bobby and Wang, Zhen and Berrill, Mark A},

  abstractNote = {Block iterative methods are extremely important as smoothers for multigrid methods, as preconditioners for Krylov methods, and as solvers for diagonally dominant linear systems. Developing robust and efficient smoother algorithms suitable for current and evolving GPU and multicore CPU systems is a significant challenge. We address this issue in the case of constant-coefficient stencils arising in the solution of elliptic partial differential equations on structured 3D uniform and adaptively refined block structured grids. Robust, highly parallel implementations of block Jacobi and chaotic block Gauss-Seidel algorithms with exact inversion of the blocks are developed using different parallelization techniques. Experimental results for NVIDIA Fermi/Kepler GPUs and AMD multicore systems are presented.},

  doi          = {10.2172/1134156},

  url          = {https://www.osti.gov/biblio/1134156},
  journal      = {},
number       = ,

  volume       = ,

  place        = {United States},

  year         = {2014},

  month        = {6}

}

Copy to clipboard

Technical Report:

https://doi.org/10.2172/1134156

Other availability

Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that may hold this item. Keep in mind that many technical reports are not cataloged in WorldCat.

Save / Share:

Export Metadata

Save to My Library

Similar records in OSTI.GOV collections:

A Block-Asynchronous Relaxation Method for Graphics Processing Units

Technical Report Antz, Hartwig ; Tomov, Stanimire ; Dongarra, Jack ; ...

In this paper, we analyze the potential of asynchronous relaxation methods on Graphics Processing Units (GPUs). For this purpose, we developed a set of asynchronous iteration algorithms in CUDA and compared them with a parallel implementation of synchronous relaxation methods on CPU-based systems. For a set of test matrices taken from the University of Florida Matrix Collection we monitor the convergence behavior, the average iteration time and the total time-to-solution time. Analyzing the results, we observe that even for our most basic asynchronous relaxation scheme, despite its lower convergence rate compared to the Gauss-Seidel relaxation (that we expected), the asynchronousmore »« less
https://doi.org/10.2172/1173288

Full Text Available
An efficient tensor transpose algorithm for multicore CPU, Intel Xeon Phi, and NVidia Tesla GPU

Journal Article Lyakh, Dmitry I. - Computer Physics Communications

An efficient parallel tensor transpose algorithm is suggested for shared-memory computing units, namely, multicore CPU, Intel Xeon Phi, and NVidia GPU. The algorithm operates on dense tensors (multidimensional arrays) and is based on the optimization of cache utilization on x86 CPU and the use of shared memory on NVidia GPU. From the applied side, the ultimate goal is to minimize the overhead encountered in the transformation of tensor contractions into matrix multiplications in computer implementations of advanced methods of quantum many-body theory (e.g., in electronic structure theory and nuclear physics). A particular accent is made on higher-dimensional tensors that typicallymore »« less
Cited by 12
https://doi.org/10.1016/j.cpc.2014.12.013

Full Text Available
An efficient mixed-precision, hybrid CPU-GPU implementation of a nonlinearly implicit one-dimensional particle-in-cell algorithm

Journal Article Chen, Guangye ; Chacon, Luis ; Barnes, Daniel C - Journal of Computational Physics

Recently, a fully implicit, energy- and charge-conserving particle-in-cell method has been developed for multi-scale, full-f kinetic simulations [G. Chen, et al., J. Comput. Phys. 230, 18 (2011)]. The method employs a Jacobian-free Newton-Krylov (JFNK) solver and is capable of using very large timesteps without loss of numerical stability or accuracy. A fundamental feature of the method is the segregation of particle orbit integrations from the field solver, while remaining fully self-consistent. This provides great flexibility, and dramatically improves the solver efficiency by reducing the degrees of freedom of the associated nonlinear system. However, it requires a particle push per nonlinearmore »« less
https://doi.org/10.1016/j.jcp.2012.04.040
Preconditioning a Newton-Krylov solver for all-speed melt pool flow physics

Journal Article Weston, Brian ; Nourgaliev, Robert ; Delplanque, Jean -Pierre ; ... - Journal of Computational Physics

n this paper, we introduce a multigrid block-based preconditioner for solving linear systems arising from a Discontinuous Galerkin discretization of the all-speed Navier-Stokes equations with phase change. The equations are discretized in conservative form with a reconstructed Discontinuous Galerkin (rDG) method and integrated with fully-implicit time discretization schemes. To robustly converge the numerically stiff systems, we use the Newton-Krylov framework with a primitive-variable formulation (pressure, velocity, and temperature), which is better conditioned than the conservative-variable form at low-Mach number. In the limit of large acoustic CFL number and viscous Fourier number, there is a strong coupling between the velocity-pressure systemmore »« less
https://doi.org/10.1016/j.jcp.2019.07.045

Full Text Available
Experiences with BoomerAMG:: A Parallel Algebraic Multigrid Solver and Preconditioner for Large Linear Systems

Conference Hensor, V E ; Yang, U M

Algebraic multigrid (AMG) is an attractive choice for solving large linear systems {Lambda}x = b on unstructured grids. While AMG is applicable as a solver for a variety of problems, its robustness may be enhanced by using it as a preconditioner for Krylov solvers, such as GMRES. The sheer size of modern problems, hundreds of millions or billions of unknowns, dictates the use of massively parallel computers. AMG consists of two phases: the setup phase, in which smaller and smaller linear systems are generated by means of linear transfer operators (interpolation and restriction); and the solve phase, which employs amore »« less
Full Text Available

Similar Records