Before starting. few Words

Transcription

1 Before starting few Words

2 Now, many Places on the Planet Earth Materials Data are produced!

3 New and Novel Materials

4 New and Novel Materials

5 New Materials from Materials Data?

6 New Materials from Materials Data? for new Materials

7 Concepts, Challenges, and Results of the (Novel Materials Discovery) Laboratory Pasquale Pavone + the entire NOMAD team Humboldt-Universität zu Berlin

8 Matthias Scheffler, FHI MPS, Berlin Kristian Thygesen Tech. U., Lyngby The NOMAD Laboratory A European Centre of Excellence Ciaran Clissmann pintail Ltd. Dublin Arndt Bode Leibniz Comp. Ctr, Garching Jose Maria Cela,BSC, Barcelona Alessandro De Vita King s Col. London Angel Rubio MPI MPSD, Hamburg Claudia Draxl Humboldt U, Berlin Risto Nieminen Aalto U. Helsinki Kimmo Koski CSC IT Center Helsinki Francesc Illas U. of Barcelona Stefan Heinzel MPS Comp. & Data, Garching Daan Frenkel U. Cambridge

9 The NOMAD Laboratory develops: Data Storage (Repository, Archive) Materials Encyclopedia Big-Data Analytics Advanced Graphics Tools for materials science and engineering

10

11 The NOMAD Scope:

12 Outline Data Storage (Repository, Archive) Materials Encyclopedia Big-Data Analytics Advanced Graphics Tools

13 Data is the raw Material of the 21st Century if you know how to use it!

14 Materials Data and their Structure Level Properties Methods Size I Atomic positions and nuclear charges, properties of free atoms, symmetry, temperature, pressure Input: definition of material gene 10 kb - 10 MB II Total energy, electron density, potential, wavefunctions, atomic forces, optimized geometry, elastic constants, etc. Density-functional theory (DFT) and ab initio molecular dynamics (MD) 10 MB - 10 TB III Excitation energies, dielectric screening, matrix elements of Coulomb interaction, etc. optical spectra, electrical conductivity, phonon spectra, thermal conductivity, etc. Many-body perturbation theory (MBPT), DF perturbation theory, ab initio MD 1 GB - 1 TB IV Efficiency of solar cell, thermoelectric figure of merit, turn-over frequency of catalyst, etc. as a function of temperature and pressure Modeling, output derived from levels I-III phenotype 10 kb - 1 MB

15 Materials Data and their Structure The amount of materials data produced on workstations, compute clusters, and supercomputers is growing exponentially. Most of it is thrown away

16 NOMAD Repository

17 : Why sharing?

18 : Why sharing?

19 : Why sharing?

20 : Why sharing?

21

22 Accepts/requests input and output files of all important codes Currently, contains more than 3,300,000 entries Tutorial video available:

23

24 Accepts/requests input and output files of all important codes Currently, contains more than 3,300,000 entries There are important codes used in computational materials science. Nomenclature, data representation, and file formats of in- and output files of these codes are different. Heterogeneity could hardly be worse.

25 The ARCHIVE

26 Conversion Layer metadata

27 : What are Metadata? Universal Vocabulary Metadata infrastructure: Is the conceptual model of our data Is format independent Describes both the data and its structure Is a posteriori (with respect to parsed codes)

28 Repository vs. Archive raw data metadata

29 Materials Encyclopedia

30 Things changed April 2017!

31

32

33

34 Big-Data Analytics

35 What is (Big-)Data Analytics?

36 Descriptors: A known Example Dmitri Mendeleev ( )

37 Predicting Crystal Structures vs. Rock salt Zincblende

38 [Å] Predicting Crystal Structures RS ZB [ev/å 2 ]

39 Advanced Graphics

40 Remote Visualization Interaction of a CO 2 with a CaO surface Video available:

41 Virtual reality

42 Further YouTube Videos

43 The NOMAD Scope Currently 57 mio. files The NOMAD Scope: Amount rapidly increasing Working on the entire DB on the search for unusual properties and phenomena Additional high-throughput calculations required for some of your needs

44 Summary TOOLS CONVERSION LAYER

45 Users Thank you for the attention!

46 Summary TOOLS CONVERSION LAYER

47 Special Thanks to: Luca Ghiringhelli FHI Berlin Georg Huhs BSC Barcelona Jungho Shin HU Berlin Claudia Draxl HU Berlin

48 Funding Acknowledgement This project has received funding from the European Union s Horizon 2020 research and innovation programme under grant agreement No