U.S. Department of EnergyOffice of Scientific and Technical Information
Plasma turbulence calculations on the Intel iPSC/860 (rx) hypercube
Abstract
One approach to improving the real-time efficiency of plasma turbulence calculations is to use a parallel algorithm. A serial algorithm used for plasma turbulence calculations was modified to allocate a radial region in each node. In this way, convolutions at a fixed radius are performed in parallel, and communication is limited to boundary values for each radial region. For a semi-implicity numerical scheme (tridiagonal matrix solver), there is a factor of 3 improvement in efficiency with the Intel iPSC/860 machine using 64 processors over a single-processor Cray-II. For block-tridiagonal matrix cases (fully implicit code), a second parallelization takes place. The Fourier components are distributed in nodes. In each node, the block-tridiagonal matrix is inverted for each of allocated Fourier components. The algorithm for this second case has not yet been optimized. 10 refs., 4 figs.
- Authors:
-
- Oak Ridge National Lab., TN (USA). Computing and Telecommunications Div.
- Oak Ridge National Lab., TN (USA)
- Albion Coll., MI (USA)
- Publication Date:
- Research Org.:
- Oak Ridge National Lab., TN (USA)
- Sponsoring Org.:
- DOE/ER
- OSTI Identifier:
- 6378468
- Report Number(s):
- CONF-910207-1
ON: DE91004773; TRN: 91-000079
- DOE Contract Number:
- AC05-84OR21400
- Resource Type:
- Conference
- Resource Relation:
- Conference: Meeting on parallel methods on large-scale structural analysis and physics applications, Hampton, VA (USA), 5-6 Feb 1991
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 70 PLASMA PHYSICS AND FUSION TECHNOLOGY; PLASMA; TURBULENCE; COMPUTER CALCULATIONS; ALGORITHMS; CHARGED-PARTICLE TRANSPORT; HYPERCUBE COMPUTERS; MHD EQUILIBRIUM; PARALLEL PROCESSING; COMPUTERS; EQUILIBRIUM; MATHEMATICAL LOGIC; PROGRAMMING; RADIATION TRANSPORT; 700103* - Fusion Energy- Plasma Research- Kinetics
Citation Formats
Lynch, V E, Carreras, B A, Drake, J B, Leboeuf, J N, and Ruiter, J R. Plasma turbulence calculations on the Intel iPSC/860 (rx) hypercube. United States: N. p., 1990.
Web.
Lynch, V E, Carreras, B A, Drake, J B, Leboeuf, J N, & Ruiter, J R. Plasma turbulence calculations on the Intel iPSC/860 (rx) hypercube. United States.
Lynch, V E, Carreras, B A, Drake, J B, Leboeuf, J N, and Ruiter, J R. 1990.
"Plasma turbulence calculations on the Intel iPSC/860 (rx) hypercube". United States. https://www.osti.gov/servlets/purl/6378468.
@article{osti_6378468,
title = {Plasma turbulence calculations on the Intel iPSC/860 (rx) hypercube},
author = {Lynch, V E and Carreras, B A and Drake, J B and Leboeuf, J N and Ruiter, J R},
abstractNote = {One approach to improving the real-time efficiency of plasma turbulence calculations is to use a parallel algorithm. A serial algorithm used for plasma turbulence calculations was modified to allocate a radial region in each node. In this way, convolutions at a fixed radius are performed in parallel, and communication is limited to boundary values for each radial region. For a semi-implicity numerical scheme (tridiagonal matrix solver), there is a factor of 3 improvement in efficiency with the Intel iPSC/860 machine using 64 processors over a single-processor Cray-II. For block-tridiagonal matrix cases (fully implicit code), a second parallelization takes place. The Fourier components are distributed in nodes. In each node, the block-tridiagonal matrix is inverted for each of allocated Fourier components. The algorithm for this second case has not yet been optimized. 10 refs., 4 figs.},
doi = {},
url = {https://www.osti.gov/biblio/6378468},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Mon Jan 01 00:00:00 EST 1990},
month = {Mon Jan 01 00:00:00 EST 1990}
}