In order to compensate for the higher cost of double double and quad double arithmetic when solving large polynomial systems, we investigate the application of NVIDIA Tesla K20C general purpose graphics processing unit. The focus on this paper is on Newton's method, which requires the evaluation of the polynomials, their derivatives, and the solution of a linear system to compute the update to the current approximation for the solution. The reverse mode of algorithmic differentiation for a product of variables is rewritten in a binary tree fashion so all threads in a block can collaborate in the computation. For double arithmetic, the evaluation and differentiation problem is memory bound, whereas for complex quad double arithmetic the problem is compute bound. With acceleration we can double the dimension and get results that are twice as accurate in about the same time.
Key words and phrases. compute unified device architecture (CUDA) double double arithmetic, differentiation and evaluation, general purpose graphics processing unit (GPU), Newton's method, least squares, massively parallel algorithm, modified Gram-Schmidt method, polynomial evaluation, polynomial differentiation, polynomial system, QR decomposition, quad double arithmetic, quality up.