Exascale revolution makes most of the existing simulation approaches and designs obsolete. The Computational Cosmology Group at The University of Chicago is developing novel approaches for the next generation cosmological simulations, such as asynchronous adaptive refinement, local time-stepping, run-time systems, multi-layered task-based paralelism on both CPU and hardware devices like GPU, etc. We are also interested in going beyond the conventional load balancing techniques and exploring new approaches for dynamically adjusting computational workload in a task-parallel way, both between separate nodes and between CPU cores and hardware devices.

Before applying for this opportunity you need to submit your online profile. Click the button below to continue.