loading...
 This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
1st IEEE Computer Society International Workshop on Cluster Computing
Lattice Field Theory on Cluster Computers: Vector- Vs. Cache-Centric Programming
Melbourne, Australia
December 02-December 03
ISBN: 0-7695-0343-8
ASCII Text
x 
 
C. Best, M. Peardon, N. Eicker, P. Uberholz, T. Lippert, K. Schilling, "Lattice Field Theory on Cluster Computers: Vector- Vs. Cache-Centric Programming," Cluster Computing, International Workshop on, pp. 312, 1st IEEE Computer Society International Workshop on Cluster Computing, 1999.
 
 
BibTex
x
 
@article{ 10.1109/IWCC.1999.810901,
author = {C. Best and M. Peardon and N. Eicker and P. Uberholz and T. Lippert and K. Schilling},
title = {Lattice Field Theory on Cluster Computers: Vector- Vs. Cache-Centric Programming},
journal ={Cluster Computing, International Workshop on},
volume = {0},
year = {1999},
issn = {1530-1885},
pages = {312},
doi = {http://doi.ieeecomputersociety.org/10.1109/IWCC.1999.810901},
publisher = {IEEE Computer Society},
address = {Los Alamitos, CA, USA},
}
 
 
RefWorks Procite/RefMan/Endnote
x 
 
TY - CONF
JO - Cluster Computing, International Workshop on
TI - Lattice Field Theory on Cluster Computers: Vector- Vs. Cache-Centric Programming
SN - 1530-1885
SP
EP
A1 - C. Best,
A1 - M. Peardon,
A1 - N. Eicker,
A1 - P. Uberholz,
A1 - T. Lippert,
A1 - K. Schilling,
PY - 1999
VL - 0
JA - Cluster Computing, International Workshop on
ER -
 
C. Best, John von Neumann Institute for Computing
M. Peardon, John von Neumann Institute for Computing
N. Eicker, University of Wuppertal
P. Uberholz, University of Wuppertal
T. Lippert, John von Neumann Institute for Computing and University of Wuppertal
K. Schilling, John von Neumann Institute for Computing and University of Wuppertal
We evaluate the possibility of moving medium-sized calculation in lattice field theory from vector supercomputers to cluster computers, namely clusters built from Alpha processors and Myrinet interconnect, and find that a medium-sized system with a performance of 10 to 20 GFlop/scan be easily and cost-effectively built from current off-the-shelf components. The performance of the algorithms is analyzed with respect to memory bandwidth problems by experiment and using a cache simulator that uses C++ operator overloading. It seems that cluster systems, while hampered by poor memory bandwidth as compared to supercomputers, might offer opportunities for some algorithms that have good locality but are not vectorizable and thus will not perform well on vector systems.
Citation:
C. Best, M. Peardon, N. Eicker, P. Uberholz, T. Lippert, K. Schilling, "Lattice Field Theory on Cluster Computers: Vector- Vs. Cache-Centric Programming," iwcc, pp.312, 1st IEEE Computer Society International Workshop on Cluster Computing, 1999
Usage of this product signifies your acceptance of the Terms of Use.