SLAC, Kavli 3rd Floor Conf. Room
The role of computation in astrophysics has grown substantially over the last decade, driven by the growth in computer power, sophistication of computational methodologies, and increasing data volumes. Over the next decade, the importance of computational astrophysics will continue to grow as computers reach exascale performance and data-intense observational surveys like the Large Synoptic Survey Telescope, Euclid, and WFIRST become available. I will present some recent efforts in computational astrophysics that push the limits of the world’s fastest supercomputers, especially by leveraging massively parallel hardware for numerical simulation. I will also show how the same hardware architectures are enabling new advances in astrophysical data analysis through deep learning. Together, numerical simulation and data analysis enabled by advanced computation will help us unlock some long standing mysteries throughout astrophysics.