SACRA-K : A Performance-portable Numerical Relativity Code with Kokkos
- 1Chinese Academy of Sciences
- 2Max-Planck-Institut für Gravitationsphysik (Albert-Einstein-Institut)
- 3Kyoto University
Abstract
We present the SimulAtor for Compact objects in Relativistic Astrophysics with Kokkos ( SACRA-K ), a performance-portable numerical relativity (NR) code ported from the Fortran code SACRA-MPI to C++ with the Kokkos library. SACRA-K retains the physics and numerical methods of SACRA-MPI , namely, a Baumgarte–Shapiro–Shibata–Nakamura (BSSN) spacetime evolution with Z4c constraint propagation and a box-in-box Berger–Oliger adaptive mesh refinement, together with the high-resolution shock-capturing scheme for the general relativistic hydrodynamics from NANASI , while gaining cross-vendor portability. We validate the port against SACRA-MPI across binary black hole, black hole neutron star, and binary neutron star (BNS) configurations. Across these tests, the waveform discrepancies are well below both the scatter among independent NR codes and the resolution dependent variation within a given code, and remain at or below the distinguishability threshold of current gravitational-wave (GW) detectors; the π symmetry is preserved exactly at the bitwise level; and the GW phase of the BNS merger exhibits second-order convergence. In the smallest test configuration, SACRA-K runs about an order of magnitude faster on the NVIDIA A100 GPU cluster or the AMD MI300A APU cluster than the Fortran SACRA-MPI on the CPU cluster, and we measure its scaling up to 256 accelerator devices.
Related
- Gravitational waves from core collapse of rotating very-massive stars: 3D numerical relativity computation
- The Bondi dipole in full numerical relativity: a self-accelerating positive-negative mass binary
- Spacetime Evolution Induced by a Massive Cosmic String in Numerical Relativity
- Considering Numerical Relativity Surrogate Models for AGN Channel Merger Dynamics
- Binary Ellis-Bronnikov wormholes in 3+1 numerical relativity: mergers, fly-bys and gravitational waves
BibTeX
@article{han2026sacrak,
title = {{SACRA-K : A Performance-portable Numerical Relativity Code with Kokkos}},
author = {Ming-Zhe Han and Kenta Kiuchi and Masaru Shibata},
journal = {The Astrophysical Journal Supplement Series},
year = {2026},
month = oct,
doi = {10.3847/1538-4365/aeac1e},
url = {https://iopscience.iop.org/article/10.3847/1538-4365/aeac1e},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - SACRA-K : A Performance-portable Numerical Relativity Code with Kokkos
AU - Ming-Zhe Han
AU - Kenta Kiuchi
AU - Masaru Shibata
PY - 2026
DA - 2026/10/27
JO - The Astrophysical Journal Supplement Series
DO - 10.3847/1538-4365/aeac1e
UR - https://iopscience.iop.org/article/10.3847/1538-4365/aeac1e
AB - We present the SimulAtor for Compact objects in Relativistic Astrophysics with Kokkos ( SACRA-K ), a performance-portable numerical relativity (NR) code ported from the Fortran code SACRA-MPI to C++ with the Kokkos library. SACRA-K retains the physics and numerical methods of SACRA-MPI , namely, a Baumgarte–Shapiro–Shibata–Nakamura (BSSN) spacetime evolution with Z4c constraint propagation and a box-in-box Berger–Oliger adaptive mesh refinement, together with the high-resolution shock-capturing scheme for the general relativistic hydrodynamics from NANASI , while gaining cross-vendor portability. We validate the port against SACRA-MPI across binary black hole, black hole neutron star, and binary neutron star (BNS) configurations. Across these tests, the waveform discrepancies are well below both the scatter among independent NR codes and the resolution dependent variation within a given code, and remain at or below the distinguishability threshold of current gravitational-wave (GW) detectors; the π symmetry is preserved exactly at the bitwise level; and the GW phase of the BNS merger exhibits second-order convergence. In the smallest test configuration, SACRA-K runs about an order of magnitude faster on the NVIDIA A100 GPU cluster or the AMD MI300A APU cluster than the Fortran SACRA-MPI on the CPU cluster, and we measure its scaling up to 256 accelerator devices.
SN - 0067-0049
SN - 1538-4365
ER -