Block Gauss–Huard algorithm with column pivoting on a hybrid CPU–GPU architecture

摘要

In this paper, different variants of the block Gauss–Huard algorithm with column pivoting andrncorrespondingly interchanging of columns are presented and implemented on a hybrid CPU–GPUrnarchitecture. The study gives numerical evidence that the algorithms yield numerical solutions as goodrnas those obtained by Gaussian elimination with partial pivoting and performance evidence that they arernmore suitable for parallel architectures. This is confirmed by solving systems of linear equations for arnset of randomly generated matrices.