Development of efficient computational kernels and linear algebra routines for out-of-order superscalar processors

  • Authors:

  • O. Bessonov;D. Fougère;B. Roux

  • Affiliations:

  • Institute for Problems in Mechanics of the Russian Academy of Sciences, 101, Vernadsky ave., 119526 Moscow, Russia;Laboratoire de Modélisation et Simulation Numérique en Mécanique (L3M), L3M-IMT, La Jetée, Technopôle de Chíteau-Gombert, 13451 Marseille Cedex 20, France;Laboratoire de Modélisation et Simulation Numérique en Mécanique (L3M), L3M-IMT, La Jetée, Technopôle de Chíteau-Gombert, 13451 Marseille Cedex 20, France

  • Venue:
  • Future Generation Computer Systems - Special issue: Parallel computing technologies
  • Year:
  • 2005

Quantified Score

Hi-index 0.00

Visualization

Abstract

We present methods for developing high performance computational kernels and dense linear algebra routines. The microarchitecture of AMD processors is analyzed with the goal to achieve peak computational rates. Approaches for implementing matrix multiplication algorithms are suggested for hierarchical memory computers. Block versions of matrix multiplication and LU-decomposition algorithms are considered. The obtained performance results for AMD processors are discussed in comparison with other approaches.