Taking Fluid Models Beyond Navier-Stokes

Amanda Peters Randles, Harvard University

In previous research, I have focused on a large-scale model coupling the fluid dynamics of blood plasma with the movement of red blood cells, from which we can start to elucidate trends and aid prognosis of cardiovascular disease based on high-resolution patient-specific data. The work presented here extends this model to accurately account for fluid flow beyond the continuum limit, such as that found in the micro-vasculature. Recent work has shown that higher-order approximations of the continuum Boltzmann equation enable simulations for fluid flows at these larger Knudsen numbers. These higher-order models, however, introduce significant communication and computational complexity to the model. I will present a performance study of the higher-order models as compared to the traditional ones, on both the IBM Blue Gene/P and Blue Gene/Q architectures. This will include a study of the tradeoffs of many optimization methods such as the use of deep halo level ghost cells that, alongside hybrid programming models, reduce the impact of extended models and enable efficient modeling of extreme regimes of computational fluid dynamics.

Abstract Author(s): Amanda Peters Randles, Vivek Kale, William Gropp, Efthimios Kaxrias