Algorithmic Aspects of High-Performance Computing for Mechanics and Physics

Santiago Badia, Universitat Politècnica de Catalunya
Victor Calo, King Abdullah University of Science and Technology
Javier Principe, Universitat Politècnica de Catalunya

The development of more powerful and energy efficient computer architectures is widening the use of large scale -high performance- computing techniques. Important increases in computational power are being achieved by many core systems with powerful accelerators such as general purpose graphics processing units (GP-GPUs). These systems also include more complex hierarchical cache/memory designs which permit to provide the needed data flow required by computing units. These new hardware designs poses a challenge for the implementation and scalability of current simulation algorithms and, at the same time, offers the an opportunity for the development of new ones. In this framework the proposed minisymposium aims to discuss algorithmic aspects of large scale computing. The topics include new developments in domain decomposition, multigrid and sparse di- rect methods, in terms of theoretical aspects, practical implementation and scalability studies. Tailored algorithms for multiphysics problems, e.g. fluid-structure interaction, porous media flows or MHD, are also welcomed.