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

Buttari, Alfredo and Dongarra, Jack and Langou, Julie and Langou, Julien and Luszczek, Piotr and Kurzak, Jakub (2007) Mixed Precision Iterative Refinement Techniques for the Solution of Dense Linear Systems. [MIMS Preprint]

PDF Mixed_Precision_Iterative.pdf Download (161kB)

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.

Item Type:	MIMS Preprint
Additional Information:	Accepted in International Journal of High Performance Computer Applications, August 2007
Subjects:	MSC 2010, the AMS's Mathematics Subject Classification > 65 Numerical analysis MSC 2010, the AMS's Mathematics Subject Classification > 68 Computer science
Depositing User:	Ms Lucy van Russelt
Date Deposited:	10 Oct 2007
Last Modified:	08 Nov 2017 18:18
URI:	https://eprints.maths.manchester.ac.uk/id/eprint/858

Actions (login required)

View Item