Fast development of dense linear algebra codes on graphics processors
Depto. de Ingenieria y Ciencia de los Computadores, Universidad Jaume I, Castellon (Spain)
IEEE International Symposium on Parallel & Distributed Processing, 2009, IPDPS 2009
@conference{zafont2009fast,
title={Fast development of dense linear algebra codes on graphics processors},
author={Zafont, M.J. and Martin, A. and Igual, F. and Quintana-Orti, ES},
booktitle={Parallel & Distributed Processing, 2009. IPDPS 2009. IEEE International Symposium on},
pages={1–8},
issn={1530-2075},
year={2009},
organization={IEEE}
}
We present an application programming interface (API) for the C programming language that facilitates the development of dense linear algebra algorithms on graphics processors applying the FLAME methodology. The interface, built on top of the NVIDIA CUBLAS library, implements all the computational functionality of the FLAME/C interface. In addition, the API includes data transference routines to explicitly handle communication between the CPU and GPU memory spaces. The flexibility and simplicity-of-use of this tool are illustrated using a complex operation of dense linear algebra: the Cholesky factorization. For this operation, we implement and evaluate all existing variants on an NVIDIA G80 processor, attaining speedups 7? compared with the CPU implementations.
January 20, 2011 by hgpu
