Abstract

Contributed Talk - Splinter CompQuantum

Wednesday, 09 September 2026, 16:15   (MW-2050)

Highlights and lessons learned from simulations at the exascale

Philipp Grete
University of Hamburg

On the path to exascale computing the landscape of computer device architectures and corresponding programming models has become much more diverse. While various low-level performance portable programming models are available, support at the application level lacks behind. To address this issue, we present the performance portable block-structured adaptive mesh refinement (AMR) framework Parthenon, which serve as the foundation for a variety of downstream astrophysics and multi-physics codes. Parthenon adopts the Kokkos programming model, and provides various levels of abstractions from multi-dimensional variables, to packages defining and separating components, to launching of parallel compute kernels. At the largest scale, a Parthenon-based hydrodynamics miniapp reaches ~92% weak scaling parallel efficiency on 9,216 nodes (73,728 logical GPUs) on Frontier at (starting from a single node). In this talk, I will highlight our performance-motivated key design decisions in developing Parthenon and recent successes in running large scale astrophysical simulations ranging from turbulence to AGN feedback. Moreover, I will share our practical experiences and challenges in scaling up with an emphasis on handling the number of concurrent messages on the interconnect, writing large output files, and post-process them for visualization – which also translate to other applications.