Performance of a plasma fluid code on the Intel parallel computers

Author

Lynch, V.E. ; Carreras, B.A. ; Drake, J.B. ; Leboeuf, J.N. ; Liewer, P.

Author_Institution

Martin Marietta Energy Systems Inc., Oak Ridge, TN, USA

fYear

1992

fDate

16-20 Nov 1992

Firstpage

286

Lastpage

293

Abstract

One approach to improving the real-time efficiency of plasma turbulence calculations is to use a parallel algorithm. A parallel algorithm for plasma turbulence calculations was tested on the Intel iPSC/860 hypercube and the Touchstone Delta machine. Using the 128 processors of the Intel iPSC/860 hypercube, a factor of 5 improvement over a single-processor CRAY-2 is obtained. For the Touchstone Delta machine, the corresponding improvement factor is 13. For plasma edge turbulence calculations, an extrapolation of the present results to the Intel σ machine gives an improvement factor close to 64 over the single-processor CRAY-2

Keywords

hypercube networks; parallel algorithms; parallel processing; physics computing; plasma flow; plasma turbulence; CRAY-2; Intel σ machine; Intel parallel computers; Touchstone Delta machine; extrapolation; iPSC/860 hypercube; parallel algorithm; plasma fluid code performance; plasma turbulence calculations; real-time efficiency; Concurrent computing; Convolutional codes; Hypercubes; Kinetic theory; Laboratories; Magnetohydrodynamics; Parallel algorithms; Partial differential equations; Plasma transport processes; Telecommunication computing;

fLanguage

English

Publisher

ieee

Conference_Titel

Supercomputing '92., Proceedings

Conference_Location

Minneapolis, MN

Print_ISBN

0-8186-2630-5

Type

conf

DOI

10.1109/SUPERC.1992.236674

Filename

236674