FAIR-DI e.V.
FAIRmat
NOMAD Laboratory
NOMAD CoE

Nov 1-3, 2021: Joint NOMAD - E-CAM Workshop

ecam-big

 

 

 

Modeling materials at realistic space and time scales via optimal exploitation of exascale computers and AI

This CECAM flagship workshop will focus on recent, important developments, addressing exascale computing and artificial intelligence (AI) methods with a specific focus on urgent topics of materials science.

The workshop hence aims at updating the audience about new and upcoming possibilities. For this, we will have lectures from top HPC centers from Europe, USA, Japan, and China. In particular, we will address how exascale computing can help materials modeling in terms of higher accuracy and precision and more controlled scale bridging. We will also discuss advancements of AI methods and exascale workflows.

These technical aspects will be presented and discussed by leading experts in different domains. By a good mix of researchers from hard and soft materials science, using first principles, classical-atomistic, coarse-grained, and continuum simulations, we will be able to explore similarities and differences in approaches to exascale and data and workflow management.

 

Format

The workshop will last five half days and comprise 21 scientific talks as well as a poster session. Talks, discussions, and poster sessions will be held as a “hybrid” of a virtual workshop and a regular face-to-face meeting.

Sessions start with an introduction by a renowned scientist, the “moderator” (15 minutes). Talks then last for 30 minutes and are followed by 15 minutes of discussion.

 

Date & Location

The Workshop takes place from November 1-3, 2021. The in-person attendees meet at CECAM HQ in Lausanne while the online participants are able to lifestream the meeting via zoom.

 
Registration

Please use the following link for registration at the workshop: https://www.cecam.org/workshop-details/1037

 

Preliminary scientific program
Workshop day 1 (November 1, 2021)

Session 1: Architectures of exascale computers and necessary coding concepts 

Moderator: Erwin Laure

15:00 - 15:15 Introduction

Erwin Laure

(Max Planck Computing and Data Facility)

15:15 - 15:45 Scientific talk 1

Pete Beckmann

(Northwestern Argonne Institute of Science and Engineering)

15:45 - 16:00 Discussion  
16:00 - 16:20 Coffee break 
16:20 - 16:50 Scientific talk 2

Satoshi Matsuka

(RIKEN)

16:50 - 17:05 Discussion  
17:05 - 17:35 Scientific talk 3

TBA

17:35 - 17:50 Discussion  
17:50 - 18:20 Scientific talk 4  TBA
18:20 - 18:35 Discussion  
18:35 - 20:30 Poster session

 

Workshop day 2 (November 2, 2021)

Session 2: Multi-scale modeling at the exascale

Moderator: Sara Bonella

09:00 - 09:15 Introduction

Sara Bonella

(CECAM EPFL)

09:15 - 09:45 Exascale challenge for workflows and multi-scale modeling

Kristian Sommer Thygesen

(Technical University of Denmark)

09:45 - 10:00 Discussion  
10:00 - 10:30 Exascale workflows

Geoffroy Hautier

(Dartmouth College)

10:30 - 10:45 Discussion  
10:45 - 11:05 Coffee break  
11:05 - 11:35 Mesoscale algorithms and codes in exascale architectures

Michael Seaton

(Hartree Center)

11:35 - 11:50 Discussion  
11:50 - 12:20 Multiscaling from CFD downwards

Petros Koumoutsakos

(Harvard University)

12:20 - 12:35 Discussion  
13:45 - 14:15 Kinetic Monte Carlo

Qian Yang

(University of Connecticut)

14:15 - 14:30 Discussion  
14:30 - 15:00 Multiscale modeling in soft and biological matter

Matej Praprotnik

(National Institute of Chemistry)

15:00 - 15:15 Discussion  

 

Session 3: Challenges in atomistic modelling

Moderator: Ignacio Pagonabarraga

15:35 - 15:50 Introduction

Ignacio Pagonabarraga

(CECAM EPFL)

15:50 - 16:20 Exascale challenges in low dimensional materials

Nicola Marzari

(EPFL)

16:20 - 16:35 Discussion  
16:35 - 17:05 Design of macromolecular products and processes from scratch

Juan de Pablo

(University of Chicago)

17:05 - 17:20 Discussion  
17:20 - 17:50 Grand canonical replica exchange MD from first principles

Luca Ghiringhelli

(Fritz Haber Institute of the Max Planck Society)

17:50 - 18:10 Discussion  

 

 Workshop day 3 (November 3, 2021)

Session 4: Artificial intelligence concepts

Moderator: Claudia Draxl

09:00 - 09:15 Introduction

Claudia Draxl

(Humboldt Universität zu Berlin & Fritz Haber Institute of the Max Planck Society)

09:15 - 09:45 Finding structure in data, identifying maps of materials properties, and detecting the "materials genes"

Matthias Scheffler

(Fritz Haber Institute of the Max Planck Society & Humboldt Universität zu Berlin)

09:45 - 10:00 Discussion  
10:00 - 10:30 Machine learned potential-energy surfaces

Roberto Car

(Princeton University)

10:30 - 10:45 Discussion  
10:45 - 11:05 Coffee Break  
11:05 - 11:35 Machine learned potential-energy surfaces

Cecilia Clementi

(Freie Universität Berlin)

11:35 - 11:50 Discussion  
11:50 - 12:20 Data-driven discovery of rare phenomena

Mario Boley

(Monash University)

12:20 - 12:35 Discussion  

 

Session 5: AI for molecular modelling 

Moderator: Kurt Kremer

14:00 - 14:15 Introduction

Kurt Kremer

(Max Planck Institute for Polymer Research)

14:15 - 14:45 AI in multiscale atomistic simulations

Gábor Csányi

(University of Cambridge)

14:45 - 15:00 Discussion  
15:00 - 15:30 Machine learning for molecular simulation

Christoph Dellago

(Universität Wien)

15:30 - 15:45 Discussion  
15:45 - 16:05 Coffee break  
16:05 - 16:35 AI for kinetic properties of coarse-grained methods

Tristan Bereau

(University of Amsterdam)

16:35 - 16:50 Discussion  
16:50 - 17:20 Big data science in porous materials

Berend Smit

(EPFL)

17:20 - 17:35 Discussion  

 * to be confirmed

The workshop is part of the CECAM flagship program!

logo