DMCA
Parallelization of structured, hierarchical adaptive mesh refinement algorithms (2000)
| Venue: | Computing and Visualization in Science |
| Citations: | 39 - 10 self |
BibTeX
@INPROCEEDINGS{Rendleman00parallelizationof,
author = {Charles A. Rendleman and Vincent E. Beckner and Mike Lijewski and William Crutchfield and John B. Bell},
title = {Parallelization of structured, hierarchical adaptive mesh refinement algorithms},
booktitle = {Computing and Visualization in Science},
year = {2000},
pages = {3--147}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
We describe an approach to parallelization of structured adaptive mesh refinement algorithms. This type of adaptive methodology is based on the use of local grids superimposed on a coarse grid to achieve sufficient resolution in the solution. The key elements of the approach to parallelization are a dynamic load-balancing technique to distribute work to processors and a software methodology for managing data distribution and communications. The methodology is based on a message-passing model that exploits the coarse-grained parallelism inherent in the algorithms. The approach is illustrated for an adaptive algorithm for hyperbolic systems of conservation laws in three space dimensions. A numerical example computing the interaction of a shock with a helium bubble is presented. We give timings to illustrate the performance of the method.
Keyphrases
hierarchical adaptive mesh refinement algorithm dynamic load-balancing technique coarse-grained parallelism inherent sufficient resolution numerical example space dimension helium bubble adaptive algorithm key element local grid structured adaptive mesh refinement algorithm coarse grid software methodology conservation law data distribution hyperbolic system message-passing model adaptive methodology
