Mixed Precision Iterative Refinement Techniques for the Solution of Dense Linear Systems - Archive ouverte HAL Access content directly
Journal Articles International Journal of High Performance Computing Applications Year : 2016

Mixed Precision Iterative Refinement Techniques for the Solution of Dense Linear Systems

Abstract

By using a combination of 32-bit and 64-bit floating point arithmetic, the performance of many dense and sparse linear algebra algorithms can be significantly enhanced while maintaining the 64-bit accuracy of the resulting solution. The approach presented here can apply not only to conventional processors but also to exotic technologies such as Field Programmable Gate Arrays (FPGA), Graphical Processing Units (GPU), and the Cell BE processor. Results on modern processor architectures and the Cell BE are presented.

Dates and versions

hal-02421121 , version 1 (20-12-2019)

Identifiers

Cite

Alfredo Buttari, Jack J. Dongarra, Julie Langou, Julien Langou, Piotr Luszczek, et al.. Mixed Precision Iterative Refinement Techniques for the Solution of Dense Linear Systems. International Journal of High Performance Computing Applications, 2016, 21 (4), pp.457-466. ⟨10.1177/1094342007084026⟩. ⟨hal-02421121⟩
57 View
0 Download

Altmetric

Share

Gmail Facebook X LinkedIn More