Abstract

Quantum-mechanical simulations have become massively used tools for scientific discovery and technological advancement: thanks to their predictive power they can suggest, accelerate, or support actual physical experiments. This is a far-reaching paradigm shift, substituting the cost- and time-scales of brick-and-mortar facilities, equipment, and personnel with those, very different, of computing engines, and aiming at understanding, predicting, or designing the properties and performance of novel or complex materials and devices with computer simulations.

I will highlight how the convergence of high-performance computing, high-throughput computing, and data analytics is driving a revolution in the effort to discover novel materials with improved properties and performance. I will discuss the software infrastructure required to create, monitor, record, reproduce, and disseminate millions of complex calculations, and illustrate its power and promise with an example dedicated to the discovery of two-dimensional materials with novel electronic, topological, or catalytic properties.