CUDA acceleration of a matrix-free Rosenbrock-K method applied to the shallow water equations

Authors:
Paul Tranquilli;Ross Glandon;Adrian Sandu
Affiliations:
Computational Science Laboratory, Science, Virginia Tech, Blacksburg, Virginia;Computational Science Laboratory, Science, Virginia Tech, Blacksburg, Virginia;Computational Science Laboratory, Science, Virginia Tech, Blacksburg, Virginia
Venue:
ScalA '13 Proceedings of the Workshop on Latest Advances in Scalable Algorithms for Large-Scale Systems
Year:
2013

Citing 5
Cited 0

Jacobian-free Newton-Krylov methods: a survey of approaches and applications

Journal of Computational Physics
Parallel Arnoldi eigensolvers with enhanced scalability via global communications rearrangement

Parallel Computing
Solving Sparse Linear Systems on NVIDIA Tesla GPUs

ICCS '09 Proceedings of the 9th International Conference on Computational Science: Part I
A structured parallel periodic arnoldi shooting algorithm for RF-PSS analysis based on GPU platforms

Proceedings of the 16th Asia and South Pacific Design Automation Conference
Accelerating the Explicitly Restarted Arnoldi Method with GPUs Using an Autotuned Matrix Vector Product

SIAM Journal on Scientific Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Many simulations of evolutionary ordinary and partial differential equations require implicit time integration methods to avoid stability restrictions on the step size. The computation and communication costs associated with solving nonlinear systems at each time step dominates the total simulation cost. Rosenbrock-Krylov (Rosenbrock-K) methods alleviate this major bottleneck by using Krylov space approximations tightly coupled with the time discretization. This work studies the performance of Rosenbrock-K methods on accelerated hardware. GPU acceleration is used to expedite computations of the semi-discrete right hand side and the linear-algebra computations in the time integration method. A novel parallelization of the Arnoldi procedure for the construction of the Krylov based approximations of the Jacobian matrix is presented. Rosenbrock-K methods' unique ability to operate almost entirely in a reduced space make them especially suitable for efficient utilization of accelerated hardware, where standard implicit approaches may lead to systems too large for device memory.