Hardware-Based Efficiency Advances in the EXA-DUNE Project

We present advances concerning efficient finite element assembly and linear solvers on current and upcoming HPC architectures obtained in the frame of the EXA-DUNE project, part of the DFG priority program 1648 Software for Exascale Computing (SPPEXA). In this project, we aim at the development of both flexible and efficient hardware-aware software components for the solution of PDEs based on the DUNE platform and the FEAST library. In this contribution, we focus on node-level performance and accelerator integration, which will complement the proven MPI-level scalability of the framework. The higher-level aspects of the EXA-DUNE project, in particular multiscale methods and uncertainty quantification, are detailed in the companion paper [Bastian el al., Advances concerning multiscale methods and uncertainty quantification in EXA-DUNE, Proceedings of the SPPEXA Symposium 2016]. 1 The EXA-DUNE Project Partial differential equations (PDEs) – often parameterised or stochastic – lie at the heart of many models for processes from science and engineering. Despite ever1 Interdisciplinary Center for Scientific Computing, Heidelberg University, Im Neuenheimer Feld 368, D-69120 Heidelberg, 2 Institute for Computational and Applied Mathematics, University of Münster, Orleans-Ring 10, D-48149 Münster, 3 Institute for Applied Analysis and Numerical Simulation, University of Stuttgart, Allmandring 5b, D-70569 Stuttgart, 4 Inst. f. Applied Mathematics, TU Dortmund, Vogelpothsweg 87, D-44227 Dortmund. 5Fraunhofer Institute for Industrial Mathematics ITWM Fraunhofer-Platz 1, D-67663 Kaiserslautern 6Institut für Mathematik, TU Clausthal-Zellerfeld, Erzstr. 1, D-38678 Clausthal-Zellerfeld