Applying OOC Techniques in the Reduction to Condensed Form for Very Large Symmetric Eigenproblems on GPUs
Rudjer Boskovic Institute, Centre for Informatics and Computing, Bijenicka street 54, 10000-Zagreb, Croatia
20th Euromicro Conference on Parallel, Distributed and Network based Processing (PDP 2012), 2012
@inproceedings{davidovic2012applying,
title={Applying OOC techniques in the reduction to condensed form for very large symmetric eigenproblems on GPUs},
author={Davidovic, D. and Quintana-Ort{i}, E.S.},
booktitle={Proceedings of the 20th Euromicro Conference on Parallel, Distributed and Network based Processing–PDP 2012},
year={2012}
}
In this paper we address the reduction of a dense matrix to tridiagonal form for the solution of symmetric eigenvalue problems on a graphics processor (GPU) when the data is too large to fit into the accelerator memory. We apply out-of-core techniques to a three-stage algorithm, carefully redesigning the first stage to reduce the number of data transfers between the CPU and GPU memory spaces, maintain the memory requirements on the GPU within limits, and ensure high performance by featuring a high ratio between computation and communication.
January 6, 2012 by hgpu
