Dominic E. Charrier

Dominic E. Charrier, Tobias Weinzierl

We present a communication- and data-sensitive formulation of ADER-DG for hyperbolic differential equation systems. Sensitive here has multiple flavours: First, the formulation reduces the persistent memory footprint. This reduces pressure on the memory subsystem. Second, the formulation realises the underlying predictor-corrector scheme with single-touch semantics, i.e., each degree of freedom is read on average only once per time step from the main memory. This reduces communication through the memory controllers. Third, the formulation breaks up the tight coupling of the explicit time stepping's algorithmic steps to mesh traversals. This averages out data access peaks. Different operations and algorithmic steps are ran on different grid entities. Finally, the formulation hides distributed memory data transfer behind the computation aligned with the mesh traversal. This reduces pressure on the machine interconnects. All techniques applied by our formulation are elaborated by means of a rigorous task formalism. They break up ADER-DG's tight causal coupling of compute steps and can be generalised to other predictor-corrector schemes.

Dominic E. Charrier, Dave A. May, Sascha M. Schnepp

Provable stable arbitrary order symmetric interior penalty discontinuous Galerkin (SIP) discretisations of variable viscosity, incompressible Stokes flow utilising $Q^2_k$--$Q_{k-1}$ elements and hierarchical Legendre basis polynomials are developed and investigated.For solving the resulting linear system, a block preconditioned iterative method is proposed. The nested viscous problem is solved by a $hp$-multilevel preconditioned Krylov subspace method. For the $p$-coarsening, a twolevel method utilising element-block Jacobi preconditioned iterations as a smoother is employed. Piecewise bilinear ($Q^2_1$) and piecewise constant ($Q^2_0$) $p$-coarse spaces are considered. Finally, Galerkin $h$-coarsening is proposed and investigated for the two $p$-coarse spaces considered. Through a number of numerical experiments, we demonstrate that utilising the $Q^2_1$ coarse space results in the most robust $hp$-multigrid method for variable viscosity Stokes flow. Using this $Q^2_1$ coarse space we observe that the convergence of the overall Stokes solver appears to be robust with respect to the jump in the viscosity and only mildly depending on the polynomial order $k$. It is demonstrated and supported by theoretical results that the convergence of the SIP discretisations and the iterative methods rely on a sharp choice of the penalty parameter based on local values of the viscosity.