A numerical model of an ocean global circulation is presented. It consists of the discretization of Laplace’s problem by means of finite differences scheme of second order that gives a linear system solved by Preconditioned Conjugate Gradient Method (PCG). In this work, we observe that the performance and the accuracy of PCG solver depend on the grid resolution and the ratio of Laplace’s coefficients. Moreover, a case study of an implementation of PCG solver with diagonal preconditioner on multi-core GPU architecture based on CUBLAS library is proposed
A CUBLAS-CUDA Implementation of PCG Method of an Ocean Circulation Model / Cuomo, Salvatore; DE MICHELE, Pasquale; Farina, Raffaele. - ELETTRONICO. - 1389:(2011), pp. 1923-1926. (Intervento presentato al convegno International Conference on Numerical Analysis and Applied Mathematics tenutosi a Halkidiki, Greece nel 19 – 25 September 2011).
A CUBLAS-CUDA Implementation of PCG Method of an Ocean Circulation Model
CUOMO, SALVATORE;DE MICHELE, PASQUALE;FARINA, RAFFAELE
2011
Abstract
A numerical model of an ocean global circulation is presented. It consists of the discretization of Laplace’s problem by means of finite differences scheme of second order that gives a linear system solved by Preconditioned Conjugate Gradient Method (PCG). In this work, we observe that the performance and the accuracy of PCG solver depend on the grid resolution and the ratio of Laplace’s coefficients. Moreover, a case study of an implementation of PCG solver with diagonal preconditioner on multi-core GPU architecture based on CUBLAS library is proposedI documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
