In this paper we evaluate two solvers for the algebraic Riccati equation –a crucial operation in control theory– based on the Newton iterative scheme for the matrix sign function. When the target architecture is a multicore server with a moderate number of cores, our Riccati solvers exploit the task-parallelism that exist in the computational kernels that compose the Newton iteration. On the other hand, for hybrid platforms that include a graphics processor (GPU), we further exploit the data-parallelism of the major kernels and off-load these computations to the graphics device. Our experimental results on a platform equipped with state-of-the-art technology from AMD, Intel and NVIDIA reveals significant acceleration factors from the GPU.
IOS Press, Inc.
6751 Tepper Drive
Clifton, VA 20124
Tel.: +1 703 830 6300
Fax: +1 703 830 2300 firstname.lastname@example.org
(Corporate matters and books only) IOS Press c/o Accucoms US, Inc.
For North America Sales and Customer Service
West Point Commons
Lansdale PA 19446
Tel.: +1 866 855 8967
Fax: +1 215 660 5042 email@example.com