GRHayL: a modern, infrastructure-agnostic, extensible library for GRMHD simulations

TL;DR

GRHayL addresses the infrastructure-lock problem in GRMHD simulations by delivering a modular, infrastructure-agnostic library of core kernels. It introduces seven gems (EOS, Con2Prim, Flux_Source, Induction, Reconstruction, Neutrinos, Atmosphere) organized around a minimal chalice to enable rapid cross-code development and microphysics integration, with a refactored IllinoisGRMHD as the exemplar. The authors validate GRHayL by implementing it in the Einstein Toolkit and BlackHoles@Home and performing cross-infrastructure comparisons against origIGM across 1D shocks, TOV stars, and binary neutron star mergers, finding comparable or improved accuracy and robustness. They also report substantial memory and performance improvements and outline a path toward GPU acceleration and expanded microphysics, underscoring GRHayL’s potential to standardize cross-infrastructure GRMHD simulations for future HPC platforms.

Abstract

Interpreting multi-messenger signals from neutron stars and black holes requires reliable general-relativistic magnetohydrodynamics (GRMHD) simulations across rapidly evolving high-performance-computing platforms, yet key algorithms are routinely rewritten within infrastructure-specific numerical-relativity codes, hindering verification and reuse. We present the General Relativistic Hydrodynamics Library (GRHayL), a modular, infrastructure-agnostic GR(M)HD library providing conservative-to-primitive recovery, reconstruction, flux/source and induction operators, equations of state, and neutrino leakage through an intuitive interface. GRHayL refactors and extends the mature IllinoisGRMHD code into reusable pointwise and stencil-wise kernels, enabling rapid development and cross-code validation in diverse frameworks, while easing adoption of new microphysics and future accelerators. We implement the same kernels in the Einstein Toolkit (Carpet and CarpetX) and BlackHoles@Home, demonstrating portability with minimal duplication. Validation combines continuous-integration unit tests with cross-infrastructure comparisons of analytic GRMHD Riemann problems, dynamical Tolman-Oppenheimer-Volkoff evolutions, and binary neutron-star mergers, showing comparable or improved behavior over legacy IllinoisGRMHD and established Einstein Toolkit codes.

Figures (16)

• Figure 1: Comparison of GRHydro, IllinoisGRMHD, and GRHayLHDX with the Balsara 0 test.
• Figure 2: Relative difference between IllinoisGRMHD and GRHayLHDX in the Balsara 0 test.
• Figure 3: Comparison of GRHydro and IllinoisGRMHD against the exact solution for the Balsara tests. The left column shows the density at the final time of the test, and the right column shows the $y$ component of the magnetic field.
• Figure 4: Top: Comparison of non-magnetized TOV initial data evolution between origIGM, IllinoisGRMHD, GRHayLHDX, and NRPyCartMHD, each at two resolutions. IllinoisGRMHD and NRPyCartMHD are equivalent until outer boundary effects are apparent, therefore we truncate the results of NRPyCartMHD before they manifest. Bottom: Comparison of convergence order between origIGM and IllinoisGRMHD. IllinoisGRMHD consistently outperforms origIGM.
• Figure 5: Comparison of non-magnetized TOV initial data evolution between origIGM and various changes made to IllinoisGRMHD during development.