Scalable approaches to long-time atomistic dynamics: a journey to the exascale

Danny Perez
Los Alamos National Laboratory
Theoretical Division T-1

Molecular dynamics (MD) is one of the most powerful tools in the arsenal of computational materials scientists. While the exponential growth in available computing power has translated into proportional gains in terms of accessible simulation sizes and accuracies, simulation times have dramatically lagged behind. I will introduce the timescale problem of MD and show how especially tailored methods can help mitigate this problem by leveraging massively-parallel platforms to simulate small systems. I will discuss how the basic ideas introduced almost 20 years ago have been gradually refined and extended with help from mathematicians, physicist, and computer scientists in order to develop a simulation capability that can leverage upcoming exascale architectures to dramatically extend simulation timescales.

Presentation (PDF File)

Back to Workshop I: Increasing the Length, Time, and Accuracy of Materials Modeling Using Exascale Computing