Using Graphics Processors for High-Performance Computation and Visualization of Plasma Turbulence

Author

Stantchev, George ; Juba, Derek ; Dorland, William ; Varshney, Amitabh

Author_Institution

Univ. of Maryland, College Park, MD

Volume

11

Issue

2

fYear

2009

Firstpage

52

Lastpage

59

Abstract

Direct numerical simulation (DNS) of turbulence is computationally intensive and typically relies on some form of parallel processing. The authors present techniques to map DNS computations to modern graphics processing units (GPUs), which are characterized by very high memory bandwidth and hundreds of SPMD (single-program-multiple-data) processors.

Keywords

data visualisation; multiprocessing systems; nuclear engineering computing; numerical analysis; parallel processing; plasma turbulence; direct numerical simulation; graphics processing units; high-performance computation; parallel processing; plasma turbulence visualization; single-program-multiple-data processors; Computational modeling; Computer architecture; Concurrent computing; Graphics; Numerical simulation; Parallel processing; Plasma devices; Plasma materials processing; Plasma simulation; Visualization; GPU programming; direct numerical simulation; graphics processing unit; plasma turbulence; scientific computing;

fLanguage

English

Journal_Title

Computing in Science & Engineering

Publisher

ieee

ISSN

1521-9615

Type

jour

DOI

10.1109/MCSE.2009.42

Filename

4784396