Efficient and Portable Krylov Eigensolver on Many Core Architectures

¹ CEA/DEN/DANS/DM2S, CEA Saclay, 91191 Gif-sur-Yvette Cedex - FRANCE
² Université de Lille1 Sciences et Technologie, 59655 Villeneuve d’Ascq Cedex – France
³ Maison de la Simulation, 91191 Gif-sur-Yvette Cedex - FRANCE

^* Corresponding Author: christophe.calvin@cea.fr

Abstract

We present in this article a highly parallel Krylov solver for large eigenvalue problems, The Explicit Restarted Arnoldi Method (ERAM). Our ERAM implementation may be executed on many core configurations, both homogeneous and heterogeneous ones, in order to take advantage of most of present and future supercomputers. From these experiments, we propose our approach for designing efficient and portable algorithms on multi-core architectures. It is based on the design of generic algorithms using TRILINOS approach and specialized implementation of elementary operations (matrix-matrix, matrix-vector, scalar product ...) on accelerators mentioned above. Some results on large sparse and dense matrices on petascale class machines using CPU and GPUs, and some first results obtained on Intel MIC processor are presented and analysed.