Core-Collapse Supernova Simulation at the Dawn of the Exascale Era

The study of core-collapse supernovae touches on a variety of questions including the origin of the chemical elements, the formation of neutron stars and black holes, the generation of gravitational waves, and the dynamics of the interstellar medium. Computational models of these events are inherently multiscale and multiphysics and have required increasing amounts of computational power as our understanding has matured. Chief among these multiphysics concerns are neutrino weak interactions with the stellar material, the theory of which is still evolving. These interactions are flavor and energy dependent and are notoriously feeble, leading to a need to employ numerically expensive kinetic representations to capture the transport of energy and lepton number. Nevertheless, significant advances in our understanding of core-collapse supernovae have been enabled by petascale computing, and many investigations now agree on the general outlines of the mechanism. I will describe some current projects, including efforts to build a new application code capable of taking advantage of exascale platforms to enable the study of a wide range of progenitor stars via simulations of adequate physical fidelity in tractable runtimes.