Using wavelets to solve Schrödinger equation in the Density Functional Theory (DFT) framework.
Simulation is playing an ever increasing role in chemistry, physics, materials science and biology. In this research project we propose to develop new methods that would significantly extend the possibilities of the simulation tools available today. Density functional theory has become a quasi standard for high accuracy calculations. Various algorithmic bottlenecks limit at present the applicability of density functional methods to large systems. Our aim is to eliminate these bottlenecks.
In particular we want to obtain a computational effort that scales only linearly with respect to system size within a systematic multiresolution basis set (wavelets) and we will implement these new algorithms on massively parallel computers. In this way we anticipate that we will be able to treat systems that are at least one order of magnitude larger than the present limit. These new developments will be incorporated into an existing and widely used electronic structure program (ABINIT), that allows us to calculate many physical and chemical properties.
Contributors
The project is based on 4 contributors:
- Dr. Thierry Deutsch, Co-ordinator, Atomistic Simulation group (L_Sim) at CEA-Grenoble, France;
- Prof. Stefan Goedecker, Computational Physics Group at the Basel University, Switzerland;
- Prof. Xavier Gonze, Unité de Physico-Chimie et de Physique des Matériaux, Université Catholique de Louvain, Belgium;
- Prof. Reinhold Schneider, Institute for Computer Science and Applied Mathematics, Christian Albrecht Universität zu Kiel (CAU), Germany.
Workpackages
The project is organized into 5 main workpackages (see also for more details the full document):
