Mixed Precision Iterative Refinement Techniques for the Solution of Dense Linear Systems
Department of Electrical Engineering and Computer Science, University Tennessee, Knoxville, Tennessee
Int. J. High Perform. Comput. Appl., Vol. 21, No. 4. (November 2007), pp. 457-466
@article{buttari2007mixed,
title={Mixed precision iterative refinement techniques for the solution of dense linear systems},
author={Buttari, A. and Dongarra, J. and Langou, J. and Langou, J. and Luszczek, P. and Kurzak, J.},
journal={International Journal of High Performance Computing Applications},
volume={21},
number={4},
pages={457},
issn={1094-3420},
year={2007},
publisher={SAGE Publications}
}
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.
December 6, 2010 by hgpu
