Internet Parallel Computing Archive

Parallel : Algorithms : Numerical

[ News | IPCA | Mirrors | Add | Search | Mail | Help | WoTUG ]

• A High Performance Parallel Strassen Implementation
  Last modified 09:58:41, 3 Jul 1995 - 1.1K - text
  See also <URL:http://www.cs.utexas.edu/users/rvdg/abstracts/SSUMMA.html>

  Abstract: In this paper, we give what we believe to be the first high performance parallel implementation of Strassen's algorithm for matrix multiplication. We show how under restricted conditions, this algorithm can be implemented plug compatible with standard parallel matrix multiplication algorithms. Results obtained on a large Intel Paragon system show a 10-20% reduction in execution time compared to what we believe to be the fastest standard parallel matrix multiplication implementation available at this time.

  Authors: Brian Grayson; Ajay Shah and Robert van de Geijn (rvdg@cs.utexas.edu). Department of Computer Sciences, University of Texas, USA.

• SUMMA: Scalable Universal Matrix Multiplication Algorithm
  Last modified 16:03:10, 1 Jun 1995 - 1.1K - text
  TR-95-13, Department of Computer Sciences, University of Texas, USA, April 1995. Available from <URL:http://www.cs.utexas.edu/users/rvdg/reports.html> as link TR-95-13. Also at <URL:http://www.netlib.org/lapack/lawns> as a LAPACK working note.

  Authors: Robert van de Geijn and Jerrell Watts (rvdg@cs.utexas.edu). Department of Computer Sciences, University of Texas, USA.