Title: Optimising a 3D multigrid algorithm for the IA-64 architecture

Authors: Markus Sturmer, Jan Treibig, Ulrich Rude

Addresses: Department of Computer Science 10 (System Simulation), University of Erlangen-Nuremberg, Cauerstrasse 6, 91058 Erlangen, Germany. ' Department of Computer Science 10 (System Simulation), University of Erlangen-Nuremberg, Cauerstrasse 6, 91058 Erlangen, Germany. ' Department of Computer Science 10 (System Simulation), University of Erlangen-Nuremberg, Cauerstrasse 6, 91058 Erlangen, Germany

Abstract: Multigrid methods are amongst the most efficient algorithms to numerically solve partial differential equations. However, standard implementations usually cannot exploit the potential of modern processors. The IA-64 architecture transferes most complexity to the software side to provide a highly superscalar design with large caches, leading to unique control over the actual execution. Exemplified on a simple multigrid solver equation in 3D and the Itanium 2 processor, we present how known performance optimisation techniques can be successfully combined. While implementation details are specific, the optimisation concept should be applicable for a wide range of numerical algorithm and CPUs.

Keywords: performance optimisation; 3D multigrid algorithms; cache blocking; streaming execution; IA-64 architecture.

DOI: 10.1504/IJCSE.2008.021110

International Journal of Computational Science and Engineering, 2008 Vol.4 No.1, pp.29 - 35

Available online: 04 Nov 2008 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article