Fully Empirical Autotuned QR Factorization For Multicore Architectures

Abstract : Tuning numerical libraries has become more difficult over time, as systems get more sophisticated. In particular, modern multicore machines make the behaviour of algorithms hard to forecast and model. In this paper, we tackle the issue of tuning a dense QR factorization on multicore architectures. We show that it is hard to rely on a model, which motivates us to design a fully empirical approach. We exhibit few strong empirical properties that enable us to efficiently prune the search space. Our method is automatic, fast and reliable. The tuning process is indeed fully performed at install time in less than one and ten minutes on five out of seven platforms. We achieve an average performance varying from 97% to 100% of the optimum performance depending on the platform. This work is a basis for autotuning the PLASMA library and enabling easy performance portability across hardware systems.
Complete list of metadatas

Cited literature [19 references]  Display  Hide  Download

https://hal.inria.fr/inria-00569514
Contributor : Emmanuel Agullo <>
Submitted on : Friday, February 25, 2011 - 12:12:42 PM
Last modification on : Thursday, January 11, 2018 - 6:22:35 AM
Long-term archiving on : Thursday, June 30, 2011 - 1:52:56 PM

Files

rr.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00569514, version 1
  • ARXIV : 1102.5328

Collections

Citation

Emmanuel Agullo, Jack Dongarra, Rajib Nath, Stanimire Tomov. Fully Empirical Autotuned QR Factorization For Multicore Architectures. [Research Report] RR-7526, INRIA. 2011, pp.22. ⟨inria-00569514⟩

Share

Metrics

Record views

380

Files downloads

805