A simple optimization strategy for the computation of 3D finite-differencing kernels on many-cores architectures is proposed. The 3D finite-differencing computation is split direction-by-direction and exploits two level of parallelism: in-core vectorization and multi-threads shared-memory parallelization. The main application of this method is to accelerate the high-order stencil computations in numerical relativity codes.
Optimization of Finite-Differencing Kernels for Numerical Relativity Applications / Alfieri, Roberto; Bernuzzi, Sebastiano; Perego, Albino; Radice, David. - ELETTRONICO. - 32:(2018), pp. 743-749. (Intervento presentato al convegno Parco 17 tenutosi a Bologna) [10.3233/978-1-61499-843-3-743].
Optimization of Finite-Differencing Kernels for Numerical Relativity Applications
Alfieri Roberto
;Bernuzzi Sebastiano;PEREGO, ALBINO;
2018-01-01
Abstract
A simple optimization strategy for the computation of 3D finite-differencing kernels on many-cores architectures is proposed. The 3D finite-differencing computation is split direction-by-direction and exploits two level of parallelism: in-core vectorization and multi-threads shared-memory parallelization. The main application of this method is to accelerate the high-order stencil computations in numerical relativity codes.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
