|
|
Detalji o izabranom predavanju:
| Seminar: | Seminar za numeričku matematiku i znan. računanje |
| Naziv predavanja: | Simulating Quantum Materials: goals, challenges and perspectives in the era of pre-exascale computing |
| Predavač: | Edoardo Di Napoli, Simulation Laboratory Quantum M |
| Vrijeme: |
29.05.2018 11:00 |
| Predavaonica: | dvorana 3. |
| Tip: |
Gost seminara |
| Opis: | Computer-based simulation in quantum Materials Science (MS) is an established field boasting 25+ years of research and the publication of more than 20,000 papers per year. Such a prolific output is possible thanks to a set of diverse computational methods implemented in a wide variety of simulation codes. These methods are based on the same set of mathematical equations which, once discretized, lead to a limited number of computational tasks. The great majority of these tasks rely on numerical linear algebra algorithms that have a complexity, based on the number of atoms simulated, which is cubic or higher. Many of these algorithms suffer from shortcomings which limit their scalability and portability on current supercomputing clusters. This problem is exacerbated by the unknown nature of supercomputing platforms capable of over one exaFLOPS. In this talk I will provide an high-level introduction to simulations of quantum materials, their known challenges and the available alternatives in tackling their methodological and computational hurdles (no previous knowledge of quantum mechanics is required :-) )
O predavaču:
Edoardo Di Napoli leads the Simulation Laboratory Quantum Materials (http://www.fz-juelich.de/ias/jsc/slqm). Since 2011 he is also an associate member of the graduate school AICES (Aachen Institute for Computational Science and Engineering) of the RWTH Aachen University (https://www.aices.rwth-aachen.de/en/).
He completed his degree at the 'La Sapienza” University in Rome after which he was the recipient of a Fulbright fellowship which lead him to pursue his graduate studies in the US. He was awarded his Ph.D. in Physics in August 2005 by the University of Texas at Austin working in the team of the Nobel laureate Steven Weinberg. From 2005 until fall 2007 Dr. Di Napoli was a postdoctoral associate at the University of North Carolina and a visiting scientist at the Perimeter Institute for Theoretical Physics in Waterloo, Canada. Between 2007 and 2009 Dr. Di Napoli started his own consulting company in the sector of renewable energy with special focus on integrated photovoltaic devices. In 2008 he worked as an executive manager for the NetIQ software corporation. At the end of 2009 I was hired as postdoctoral researcher by AICES graduate school. From 2011 until 2013 he was the recipient of the WolkswagenStiftung fellowship "Computational Science".
Dr. Di Napoli interests range from mathematical modeling to numerical linear algebra and computational physics with a special focus of enabling high-performance large scale quantum mechanical simulations. His work is quite interdisciplinary in nature encompassing technical knowledge in parallel programming, to algorithm optimization applied to code development and code modernization. He is the author of several publications at the interface between numerical linear algebra and scientific computing. Besides the involvement in numerical methods and algorithm optimization, he leads the new SimLab towards the development of new methods for the simulation of materials for energy applications, as well as the use of data mining techniques for the prediction of macroscopic properties of quantum materials. |
|
|
