JARA-HPC Simulation Laboratory “ab initio Methods in Chemistry and Physics”
On November 8, the first workshop of the JARA-HPC Simulation Laboratory “ab initio Methods in Chemistry and Physics” (SLai) took place in the JSC Rotunda Hall. The workshop constituted the opening event introducing SLai to the scientific community conducting research in solid-state physics and quantum chemistry. The event brought together scientists, from the main institutes within the Jülich Aachen Research Alliance (JARA), participating in the Simulation Laboratory.
By sharing their work and scientific expertise the participants helped defining a platform allowing the mutual collaboration and prioritization of the Lab's activities. Short informal talks were given by speakers from each of the main participating partners institutes and cross-sectional groups. Each talk was followed by an extended Q&A period favoring a constructive and informal interaction between the speakers and the audience. About fifty participants from RWTH and the Forschungszentrum Jülich attended the workshop.
Dr. D. Wortmann (PGI-1, FZJ) – Slides
S. Maintz (IAC, RWTH) – Slides
Prof. Dr. R. Mazzarello (Physics, RWTH)
Dr. P. Baumeister (PGI-1, FZJ) – Slides
Prof. P. Bientinesi, PhD (AICES, RWTH) – Slides
Dr.-Ing. B. Mohr (JSC, FZJ) - Slides
Dr. T. Müller (JSC, FZJ) – Slides
Prof. A. L. Tchougréeff, PhD (IAC, RWTH) – Slides
E. Rabel (PGI-1, FZJ) – Slides
Dr. E. Di Napoli (JARA-HPC) – Slides
One of the aims of the workshop was to maximize multidisciplinary cooperation. As a first step in this direction, the workshop promoted a strategic partnership between its staff and several groups of RWTH Aachen and Forschungszentrum Jülich (FZJ) with very distinct competencies.
- HPC Numerical Linear Algebra
- High-Performance Automatic Computing group (AICES, RWTH)
- Mathematical Methods and Algorithms cross-sectional team (JSC, FZJ)
- Parallel Programming and Efficiency
- Performance Analysis cross-sectional team (JSC, FZJ)
- Parallel Efficiency cross-sectional group (JARA-HPC)
- Quantum Chemistry
- Solid-State and Quantum Chemistry Chair (IAC, RWTH)
- Density Functional Theory
- Quantum Theory of Materials group (PGI-1, FZJ)
- Theoretical Nanoelectronics group (Physics, RWTH)
Current and Future Challenges
Ab initio methods are implemented in a rich spectrum of electronic structure methods to tackle various scientific problems (e.g. band gaps, structural relaxation, magnetic properties, phonon transport, chemical bonds, crystallization). Driving the progress in these methods requires an increasing cooperation within a multidisciplinary community that probably no working groups worldwide can afford. Codes usually are the result of significant program implementations performed over many years. Each one focuses on different physical aspects like, for example, finite-size versus periodic systems or molecular versus solids compounds, etc.
An important result of the workshop was to identify the medium- and long-term challenges that should be part of Slai future activities:
- simulating larger physical systems;
- dealing with the increasing complexity and proliferation of hybrid functionals;
- parallelization and portability of ab initio codes on cpu and gpu clusters;
- increasing the precision and reliability of ab initio simulations;
- invest in the development of multi-level / multi-scale models;
- invest on complexity reduction and scalability;
- development of HPC numerical linear algebra kernels;
- unification of input/output interfaces;
- developing novel charge density mixing strategies as to improve the robustness of DFT SCF convergence;
- develop a common framework for the visualization of output.
Future collaborations between participants of the workshop will be reported and listed here.