Chana: A GPU-Accelerated Adaptive-Mesh Code for General Relativistic Magnetohydrodynamics with Multigroup M1 Neutrino Transport
Abstract
This paper presents Chana, a GPU-accelerated, block-structured adaptive-mesh code for general relativistic magnetohydrodynamics, multigroup M1 neutrino transport, and dynamical spacetime evolution using the Z4c formulation. Each physical module uses independently selected spatial discretizations and ghost-zone widths to balance accuracy and computational cost. Stiff radiation-matter interactions are treated implicitly, with Jacobians calculated analytically from the chain-rules. The code is written in C++20 and it uses MPI and Kokkos 5 for distributed-memory and performance-portable parallelism. The code is validated through a number of GRMHD, radiation transport, and dynamic spacetime tests. For a representative core-collapse supernova configuration with three neutrino species and 12 energy groups per species, the code achieves approximately 1.1 × 106 cell updates per GPU second on NVIDIA A100 GPUs using four-stage RK4 integration. Weak-scaling efficiency remains above 85\% up to 256 GPUs on Perlmutter.
Related
- FUGUE: A New GPU-accelerated General Relativistic Magnetohydrodynamics Code with Weak-Field Multipole Self-Gravity and Quasi-Stationary Kerr Black Hole Evolution
- Relativistic Magnetohydrodynamics from Myers-Pospelov Lorentz-Violating Electrodynamics
- Linear causality and stability constraints on relativistic second-order magnetohydrodynamics
- An HLLD Implementation for General Relativistic Magnetohydrodynamics in AthenaK
- Extending the infrastructure of the BAM code towards resistive general-relativistic magnetohydrodynamics: tests and first applications
BibTeX
@misc{wang2026chana,
title = {{Chana: A GPU-Accelerated Adaptive-Mesh Code for General Relativistic Magnetohydrodynamics with Multigroup M1 Neutrino Transport}},
author = {Tianshu Wang},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.07267},
eprint = {2610.07267},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.07267},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Chana: A GPU-Accelerated Adaptive-Mesh Code for General Relativistic Magnetohydrodynamics with Multigroup M1 Neutrino Transport
AU - Tianshu Wang
PY - 2026
DA - 2026/10/20
PB - arXiv
DO - 10.48550/arxiv.2610.07267
UR - https://arxiv.org/abs/2610.07267
AB - This paper presents Chana, a GPU-accelerated, block-structured adaptive-mesh code for general relativistic magnetohydrodynamics, multigroup M1 neutrino transport, and dynamical spacetime evolution using the Z4c formulation. Each physical module uses independently selected spatial discretizations and ghost-zone widths to balance accuracy and computational cost. Stiff radiation-matter interactions are treated implicitly, with Jacobians calculated analytically from the chain-rules. The code is written in C++20 and it uses MPI and Kokkos 5 for distributed-memory and performance-portable parallelism. The code is validated through a number of GRMHD, radiation transport, and dynamic spacetime tests. For a representative core-collapse supernova configuration with three neutrino species and 12 energy groups per species, the code achieves approximately 1.1 × 10^6 cell updates per GPU second on NVIDIA A100 GPUs using four-stage RK4 integration. Weak-scaling efficiency remains above 85\% up to 256 GPUs on Perlmutter.
ER -