ForestClaw: Hybrid forest-of-octrees AMR for hyperbolic conservation laws

Abstract. We present a new hybrid paradigm for parallel adaptive mesh refinement (AMR) that combines the scalability and lightweight architecture of tree-based AMR with the computational efficiency of patch-based solvers for hyperbolic con-servation laws. The key idea is to interpret each leaf of the AMR hierarchy as one uniform compute patch in Rd with md degrees of freedom, where m is customarily between 8 and 32. Thus, computation on each patch can be optimized for speed, while we inherit the flexibility of adaptive meshes. In our work we choose to in-tegrate with the p4est AMR library since it allows us to compose the mesh from multiple mapped octrees and enables the cubed sphere and other nontrivial multi-block geometries. We describe aspects of the parallel implementation and close with scalings for both MPI-only and OpenMP/MPI hybrid runs, where the largest MPI run executes on 16,384 CPU cores