A Caltech Library Service

Lattice QCD: Commercial vs. Home-Grown Parallel Computers

Baillie, Clive F. (1990) Lattice QCD: Commercial vs. Home-Grown Parallel Computers. In: Proceedings of the Fifth Distributed Memory Computing Conference, 1990. IEEE , Piscataway, NJ, pp. 397-402. ISBN 0-8186-2113-3.

[img] PDF - Published Version
See Usage Policy.


Use this Persistent URL to link to this item:


Numerical simulations of Lattice QCD have been performed on practically every computer, since its inception almost twenty years ago. Lattice QCD is an ideal problem for parallel machines as it can be easily domain decomposed. In fact, the urge to simulate QCD has led to the development of several home-grown parallel “QCD machines” , in particular the Caltech Cosmic Cube, the Columbia Machine, IBM’s GF11, APE in Rome and the Fermilab Machine. These machines were built because, at the time, there were no commercial parallel computers fast enough. Today however the situation has changed with the advent of computers like the Connection Machine 2 and the Ncube 2. Herein, I shall explain why Lattice QCD is such a parallel problem and compare two large-scale simulations of it - one on the commercial Connection Machine and the other on the latest Caltech/JPL hypercube.

Item Type:Book Section
Related URLs:
URLURL TypeDescription
Additional Information:© 1990 IEEE.
Record Number:CaltechAUTHORS:20170627-160823516
Persistent URL:
Official Citation:C. F. Baillie, "Lattice QCD: Commercial vs. Home-Grown Parallel Computers," Proceedings of the Fifth Distributed Memory Computing Conference, 1990., 1990, pp. 397-402. doi: 10.1109/DMCC.1990.555412
Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:78620
Deposited By: Kristin Buxton
Deposited On:27 Jun 2017 23:42
Last Modified:15 Nov 2021 17:41

Repository Staff Only: item control page