A floating point engine for lattice gauge calculations

D. Husby, R. Atac, A. Cook, J. Deppe, M. Fischler, I. Gaines, T. Nash, T. Pham, T. Zmuda, E. Eichten, P. Mackenzie, G. Hockney, H. B. Thacker, D. Toussaint

Research output: Contribution to journalArticle

1 Scopus citations

Abstract

The latest in low cost computing solutions from the Fermilab Advanced Computer Program is targeted at Lattice Gauge theory calculations and delivers supercomputer performance at a fraction of the cost. A typical system with 256 processors, 2.5 Gigabytes of memory, and 64 Gigabytes of on-line tape storage, delivers a peak performance of 5 billion floating point operations per second. The programming environment, Canopy, provides a comprehensive, hardware independent, distributed processing platform from within the more familiar environments of FORTRAN, C, and UNIX. This paper describes the individual processing elements of the system and gives a brief description of the Canopy software.

Original languageEnglish (US)
Pages (from-to)734-737
Number of pages4
JournalIEEE Transactions on Nuclear Science
Volume36
Issue number1
DOIs
StatePublished - Feb 1989
Externally publishedYes

ASJC Scopus subject areas

  • Nuclear and High Energy Physics
  • Nuclear Energy and Engineering
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'A floating point engine for lattice gauge calculations'. Together they form a unique fingerprint.

  • Cite this

    Husby, D., Atac, R., Cook, A., Deppe, J., Fischler, M., Gaines, I., Nash, T., Pham, T., Zmuda, T., Eichten, E., Mackenzie, P., Hockney, G., Thacker, H. B., & Toussaint, D. (1989). A floating point engine for lattice gauge calculations. IEEE Transactions on Nuclear Science, 36(1), 734-737. https://doi.org/10.1109/23.34538