International Journal of Parallel Programming

, Volume 33, Issue 2, pp 165–181

Empirical Optimization for a Sparse Linear Solver: A Case Study

  • Yoon-Ju Lee
  • Pedro C. Diniz
  • Mary W. Hall
  • Robert Lucas
Article

DOI: 10.1007/s10766-005-3581-7

Cite this article as:
Lee, Y., Diniz, P.C., Hall, M.W. et al. Int J Parallel Prog (2005) 33: 165. doi:10.1007/s10766-005-3581-7

Abstract

This paper describes initial experiences with semi-automated performance tuning of a sparse linear solver in LS-DYNA, a large, widely used engineering application. Through a collection of tools supporting empirical optimization, we alleviate the burden of performance tuning for mapping today’s sophisticated engineering software to increasingly complex hardware platforms. We describe a tool that automatically isolates code segments to create benchmark subsets for the purposes of performance tuning. We present a collection of automatically generated empirical results that demonstrate the sensitivity of the application’s performance to optimization parameters. Through this case study, we demonstrate the importance of developing automatic performance tuning support for performance-sensitive applications.

Keywords

Memory hierarchy optimizationperformance tuning

Copyright information

© Springer Science+Business Media, Inc. 2005

Authors and Affiliations

  • Yoon-Ju Lee
    • 1
  • Pedro C. Diniz
    • 1
  • Mary W. Hall
    • 1
  • Robert Lucas
    • 1
  1. 1.Information Sciences InstituteUniversity of Southern CaliforniaMarina del ReyUSA