Algebraic Multigrid on GPU

Kód výsledku v IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21340%2F13%3A00210924" target="_blank" >RIV/68407700:21340/13:00210924 - isvavai.cz</a>
Výsledek na webu
—
DOI - Digital Object Identifier
—

Jazyk výsledku
angličtina
Název v původním jazyce
Algebraic Multigrid on GPU
Popis výsledku v původním jazyce
This article deals with the method of algebraic multigrid and its parallelization on GPU. Algebraic multigrid is a sparse matrix iterative solver, which finds the solution by solving also restricted versions of the original problem. The main difference from more widely known geometric multigrid is that it can create the restricted problems without any knowledge about the matrix origin and therefore it can be used for larger range of problems. The article farther presents possibilities how to parallelizethis algorithm on multicore CPU architecture and on GPU. Finally it also shows speedup obtained by the GPU parallelization.
Název v anglickém jazyce
Algebraic Multigrid on GPU
Popis výsledku anglicky
This article deals with the method of algebraic multigrid and its parallelization on GPU. Algebraic multigrid is a sparse matrix iterative solver, which finds the solution by solving also restricted versions of the original problem. The main difference from more widely known geometric multigrid is that it can create the restricted problems without any knowledge about the matrix origin and therefore it can be used for larger range of problems. The article farther presents possibilities how to parallelizethis algorithm on multicore CPU architecture and on GPU. Finally it also shows speedup obtained by the GPU parallelization.

Rok uplatnění
2013
Kód důvěrnosti údajů
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Podobné výsledky(10)