# Exposing Fine-Grained Parallelism in Algebraic Multigrid Methods

@article{Bell2012ExposingFP, title={Exposing Fine-Grained Parallelism in Algebraic Multigrid Methods}, author={Nathan Bell and Steven Dalton and Luke N. Olson}, journal={SIAM J. Sci. Comput.}, year={2012}, volume={34} }

Algebraic multigrid methods for large, sparse linear systems are a necessity in many computational simulations, yet parallel algorithms for such solvers are generally decomposed into coarse-grained...

#### Figures, Tables, and Topics from this paper

#### 160 Citations

A Dynamic Pattern Factored Sparse Approximate Inverse Preconditioner on Graphics Processing Units

- Computer Science, Mathematics
- SIAM J. Sci. Comput.
- 2019

One of the most time-consuming tasks in the procedures for the numerical study of PDEs is the solution to linear systems of equations. To that purpose, iterative solvers are viewed as a promising a...

Multi-GPU Acceleration of Algebraic Multigrid Preconditioners

- Computer Science
- 2016

Large linear system are solved which result from electrostatic field problems after discretization with the Finite Element Method with a multi-GPU implementation of Krylov subspace methods with an algebraic multigrid preconditioners. Expand

Communication-Avoiding Optimization of Geometric Multigrid on GPUs

- 2012

Multigrid methods are widely used to accelerate the convergence of iterative solvers for linear systems in a number of different application areas. In this report, we explore communication-avoiding… Expand

Data parallel algebraic multigrid

- Computer Science
- 2015

This dissertation proposes and analyzes a novel strategy to accelerate graph Laplacian eigenvector computations by combining iterative methods, namely blocked Lanczos, with breadth first search operations on graphs and demonstrates notable performance improvement compared with generic eigensolvers. Expand

A comparison of algebraic multigrid preconditioners using graphics processing units and multi-core central processing units

- Computer Science
- HiPC 2012
- 2012

The influence of multi-core central processing units and graphics processing units on several algebraic multigrid methods is investigated in this work. Different performance metrics traditionally… Expand

Parallel Unsmoothed Aggregation Algebraic Multigrid Algorithms on GPUs

- Computer Science, Mathematics
- 2013

A parallel algebraic multigrid method for isotropic graph Laplacian problems on multicore graphical processing units (GPUs) based on the aggregation framework is designed and implemented. Expand

Accelerating algebraic multigrid solvers on NVIDIA GPUs

- Computer Science
- Comput. Math. Appl.
- 2015

Numerical experiments show that the solution phase of the GPU-based algebraic solvers is up to 13 times faster than that of the sequential CPU- based algebraic multigrid solvers on the workstation. Expand

GPU Acceleration of Algebraic Multigrid Preconditioners for Discrete Elliptic Field Problems

- Physics
- IEEE Transactions on Magnetics
- 2014

The simulation of coupled electromagnetic/thermal problems with high resolution requires efficient numerical schemes. High-performance computing languages like CUDA help in unlocking the massively… Expand

Multi-GPU Acceleration of Algebraic Multi-Grid Preconditioners for Elliptic Field Problems

- Computer Science
- IEEE Transactions on Magnetics
- 2015

A multi-graphic processing unit (GPU) implementation of Krylov sub-space methods with algebraic multi-grid preconditioners is proposed to solve large linear systems stemming from finite element or finite difference discretizations of elliptic problems as they occur, e.g., in electrostatics. Expand

AMGCL: an Efficient, Flexible, and Extensible Algebraic Multigrid Implementation

- Computer Science, Mathematics
- Lobachevskii Journal of Mathematics
- 2019

The paper presents AMGCL -- an opensource C++ library implementing the algebraic multigrid method (AMG) for solution of large sparse linear systems of equations, usually arising from discretization… Expand

#### References

SHOWING 1-10 OF 67 REFERENCES

A Survey of Parallelization Techniques for Multigrid Solvers

- Computer Science
- Parallel Processing for Scientific Computing
- 2006

This paper surveys the techniques that are necessary for constructing computationally efficient parallel multigrid solvers, including traditional spatial partitioning and more novel additive multilevel methods. Expand

Using GPUs to improve multigrid solver performance on a cluster

- Computer Science
- Int. J. Comput. Sci. Eng.
- 2008

This paper explores the coupling of coarse and fine-grained parallelism for Finite Element (FE) simulations based on efficient parallel multigrid solvers by addressing the issue of limited precision on GPUs by applying a mixed precision, iterative refinement technique. Expand

Sparse matrix solvers on the GPU: conjugate gradients and multigrid

- Computer Science
- ACM Trans. Graph.
- 2003

This work implemented two basic, broadly useful, computational kernels: a sparse matrix conjugate gradient solver and a regular-grid multigrid solver for high-intensity numerical simulation of geometric flow and fluid simulation on the GPU. Expand

Parallel Smoothed Aggregation Multigrid : Aggregation Strategies on Massively Parallel Machines

- Computer Science
- ACM/IEEE SC 2000 Conference (SC'00)
- 2000

This paper considers parallelization of the smoothe aggregation multigrid methods, and discusses three different parallel aggregation algorithms an illustrates the advantages an disadvantages of each variant in terms of parallelism an convergence. Expand

Algebraic Multigrid

- 2007

In computer simulation of particular systems (e.g. in physics, biology, economics), partial differential equations (PDE) are solved numerically, mainly by applying Finite Element (FE) or Finite… Expand

BoomerAMG: a parallel algebraic multigrid solver and preconditioner

- Mathematics
- 2002

Driven by the need to solve linear systems arising from problems posed on extremely large, unstructured grids, there has been a recent resurgence of interest in algebraic multigrid (AMG). AMG is… Expand

A multigrid solver for boundary value problems using programmable graphics hardware

- Computer Science
- HWWS '03
- 2003

A system, built on programmable graphics hardware, able to solve a variety of partial differential equations with complex boundary conditions, and implements the multigrid method, a fast and popular approach to solving large boundary value problems. Expand

Parallel Smoothed Aggregation Multigrid : Aggregation Strategies on Massively Parallel Machines

- 2000

ernment retains a non exclui~ royalty-free license to publish or m. produce the published form ef thfe contribution, br allow others to do SQ for United StSteS GoW3rnmeni ~ Algebraic multigrid… Expand

A new perspective on strength measures in algebraic multigrid

- Mathematics, Computer Science
- Numer. Linear Algebra Appl.
- 2010

A framework to interpret and clarify existing measures through differential equations is presented and this framework leads to a new procedure for making pointwise strength-of-connection decisions that combines knowledge of local algebraically smooth error and of the local behavior of interpolation. Expand

Robustness and Scalability of Algebraic Multigrid

- Mathematics, Computer Science
- SIAM J. Sci. Comput.
- 2000

Some of the situations in which standard AMG does not work well are shown and the current directions taken by AMG researchers to alleviate these difficulties are indicated. Expand