HTR solver: An open-source exascale-oriented task-based multi-GPU high-order code for hypersonic aerothermodynamics

Abstract

In this study, the open-source Hypersonics Task-based Research (HTR) solver for hypersonic aerothermodynamics is described. The physical formulation of the code includes thermochemical effects induced by high temperatures (vibrational excitation and chemical dissociation). The HTR solver uses high-order TENO-based spatial discretization on structured grids and efficient time integrators for stiff systems, is highly scalable in GPU-based supercomputers as a result of its implementation in the Regent/Legion stack, and is designed for direct numerical simulations of canonical hypersonic flows at high Reynolds numbers. The performance of the HTR solver is tested with benchmark cases including inviscid vortex advection, low- and high-speed laminar boundary layers, inviscid one-dimensional compressible flows in shock tubes, supersonic turbulent channel flows, and hypersonic transitional boundary layers of both calorically perfect gases and dissociating air. Program summaryProgram Title: Hypersonics Task-based Research solverProgram Files doi:http://dx.doi.org/10.17632/9zsxjtzfr7.1Licensing provisions: BSD 2-clauseProgramming language: RegentNature of problem: This code solves the Navier–Stokes equations at hypersonic Mach numbers including finite-rate chemistry for air dissociation along with multicomponent transport. The solver is designed for direct numerical simulations (DNS) of transitional and turbulent hypersonic turbulent flows at high enthalpies, and accounts for thermochemical effects such as vibrational excitation and chemical dissociation.Solution method: This code uses a low-dissipation sixth-order targeted essentially non-oscillatory (TENO) scheme for the spatial discretization of the conservation equations on Cartesian stretched grids. The time advancement is performed either with an explicit method, when the chemistry is slow and therefore does not introduce additional stiffness in the integration, or with an operator-splitting method that integrates the chemical production rates with an implicit discretization.Additional comments: The HTR solver builds on the runtime Legion [1] and is written in the programming language Regent [2] developed at Stanford University. Instructions for the installation of the components are provided in the README file enclosed with the HTR solver and in the Legion repository [1].