Fresh from the NVIDIA Numeric Libraries Team, a white paper illustrating the use of the CUSPARSE and CUBLAS libraries to achieve a 2x speedup of incomplete-LU- and Cholesky-preconditioned interative methods. The paper focuses on the Bi-Conjugate Gradient and stabilized Conjugate Gradient iterative methods that can be used to solve large sparse nonsymmetric and symmetric positive definite linear systems, respectively. The paper also comments on the parallel sparse triangular solver, which is an essential building block in these algorithms.
Read the technical review on our NVIDIA Research Site
Download the full white paper.
Download the webinar recording discussing this white paper: www.nvidia.com/webinars
Meet the NVIDIA CUDA Numeric Team and ask live questions during a special Live Q&A Webinar on July 27th.