Workshop II: Scale-Bridging Materials Modeling at Extreme Computational Scales

April 17 - 21, 2023

Overview

Quantitatively predicting the properties of “real” structural materials is an extremely challenging endeavor, as macroscale properties depend on characteristics of the material at every scale, from nanometers (short range order, point defects, etc.), to micrometers (grain size, extended defects, etc.), to centimeters (texture, etc.). Similarly, relevant timescales range from atomic vibrations (picoseconds) to microstructural evolution times (hours to years). This extreme breadth of size and time scales makes the accurate simulations with fully-resolved atomic-scale tools (e.g., molecular dynamics) hopeless. Practical solutions must therefore rely on scale-bridging approaches that systematically upscale the lower scale physics into computationally tractable higher-scale constructs. The premise of this workshop is that extreme-scale computing can breathe new life into the field of multiscale modeling by addressing the problems identified above with brute force computing. This workshop will focus on new mathematical approaches to multiscale/multiphysics modeling, with a particular emphasis on the many theoretical and numerical challenges faced at the exascale. The goal is to bring together specialists in a range of massively parallel algorithms and researchers interested in improving the scalability of current techniques.

Topics that will be covered in this workshop include:

  • Scalable mathematical formulations for concurrent multi-scale simulations at massive scales.
  • Automated derivation of coarse-grained model from massive-scale simulations.
  • Uncertainty-quantification-driven parameterization of multiscale models.
  • Design of scalable multiscale solvers.
  • Implementation of concurrent multiscale models at massive scales.

 

This workshop will include a poster session; a request for posters will be sent to registered participants in advance of the workshop.

Program Flyer

Organizing Committee

Irene Beyerlein (University of California, Santa Barbara (UCSB))
Thomas Hudson (University of Warwick)
Thomas Swinburne (Centre National de la Recherche Scientifique (CNRS))
Anna Vainchtein (University of Pittsburgh)