Pith. sign in

REVIEW 18 cited by

AthenaK: A Performance-Portable Version of the Athena++ AMR Framework

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2409.16053 v1 pith:AO7P7VIR submitted 2024-09-24 astro-ph.IM

AthenaK: A Performance-Portable Version of the Athena++ AMR Framework

classification astro-ph.IM
keywords athenakframeworkhydrodynamicsapplicationsathenadescribegpusimplemented
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

We describe AthenaK: a new implementation of the Athena++ block-based adaptive mesh refinement (AMR) framework using the Kokkos programming model. Finite volume methods for Newtonian, special relativistic (SR), and general relativistic (GR) hydrodynamics and magnetohydrodynamics (MHD), and GR-radiation hydrodynamics and MHD, as well as a module for evolving Lagrangian tracer or charged test particles (e.g., cosmic rays) are implemented using the framework. In two companion papers we describe (1) a new solver for the Einstein equations based on the Z4c formalism and (2) a GRMHD solver in dynamical spacetimes also implemented using the framework, enabling new applications in numerical relativity. By adopting Kokkos, the code can be run on virtually any hardware, including CPUs, GPUs from multiple vendors, and emerging ARM processors. AthenaK shows excellent performance and weak scaling, achieving over one billion cell updates per second for hydrodynamics in three-dimensions on a single NVIDIA Grace Hopper processor and with a typical parallel efficiency of 80% on 65536 AMD GPUs on the OLCF Frontier system. Such performance portability enables AthenaK to leverage modern exascale computing systems for challenging applications in astrophysical fluid dynamics, numerical relativity, and multimessenger astrophysics.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 18 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. The Origin of Da Scaling: Suppressed Cooling in Fast-Cooling Mixing Layers

    astro-ph.GA 2026-06 unverdicted novelty 7.0

    Ram-pressure suppression of interface folding in fast-cooling TRMLs produces Ė_cool ∝ Da^{1/4} scaling.

  2. Ceci n'est pas une Couche de M\'elange: The Meaning of Resolved Turbulent Radiative Mixing

    astro-ph.GA 2026-06 unverdicted novelty 7.0

    The resolution independence of cooling rates in TRML simulations is an artifact of opposing numerical effects with no physical basis, and phase structure converges only when the turbulent Field length (where eddy turn...

  3. Exact Mass Conservation in Binary Neutron Star Merger Simulations

    gr-qc 2026-05 unverdicted novelty 7.0

    A rescaling algorithm for artificial atmospheres achieves exact mass and electron number conservation to round-off precision in binary neutron star merger simulations.

  4. Trapping, Irregular Waveforms, and Efficient Radiation in Ultra-relativistic Black Hole Encounters

    gr-qc 2026-04 unverdicted novelty 7.0

    Ultra-relativistic black hole flybys can radiate over 65% of their energy in gravitational waves via irregular waveforms caused by radiation trapping and lensing, without coalescence.

  5. Turbulence Mode Decomposition and Anisotropy in Magnetically Dominated Collisionless Plasmas

    physics.plasm-ph 2026-04 unverdicted novelty 7.0

    In magnetically dominated relativistic collisionless plasmas, Alfvén and slow modes follow anisotropic Goldreich-Sridhar scaling, fast modes are isotropic with larger kinetic energy fraction, and dynamic alignment wea...

  6. GRACE: An Open-Source Framework for GPU-Accelerated Numerical Relativity

    gr-qc 2026-07 accept novelty 6.0

    GRACE is a validated, open-source, Kokkos+p4est GPU-portable framework that evolves ideal GRMHD with constrained transport self-consistently coupled to Z4c Einstein equations on fixed or adaptive meshes.

  7. Clumps in a Cocoon: Geometry and Mixing Set the Universal X-ray to H$\alpha$ Surface Brightness Ratio

    astro-ph.GA 2026-06 unverdicted novelty 6.0

    3D wind-tunnel simulations in the χ~10^3 regime show clump-cocoon geometry sets SB_X/SB_Hα~3, with Hα fraction fixed by atomic physics and X-ray fraction set by residence time in the X-ray band that scales inversely w...

  8. Effects of magnetically driven shocks on nucleosynthesis and kilonovae from neutron star mergers

    astro-ph.HE 2026-05 unverdicted novelty 6.0

    Magnetically driven shocks from neutron star merger remnants can reheat ejecta to nuclear statistical equilibrium, alter r-process yields, and produce observable changes in kilonova color and light curves.

  9. SFUMATO#: A GPU-accelerated code for self-gravitational radiation hydrodynamics simulation with adaptive mesh refinement

    astro-ph.GA 2026-04 unverdicted novelty 6.0

    SFUMATO# implements GPU-accelerated AMR for self-gravitational radiation hydrodynamics with new linearized implicit solvers for non-equilibrium chemistry and thermal evolution including dust.

  10. SFUMATO#: A GPU-accelerated code for self-gravitational radiation hydrodynamics simulation with adaptive mesh refinement

    astro-ph.GA 2026-04 accept novelty 6.0

    SFUMATO# is a multi-GPU AMR code for self-gravitational radiation hydrodynamics featuring new linearized implicit solvers for non-equilibrium chemistry and thermal evolution, validated on test problems with measured s...

  11. Diffusion of PeV Cosmic Rays in the Turbulent and Multiphase Interstellar Medium

    astro-ph.HE 2026-04 unverdicted novelty 6.0

    Multiphase ISM simulations show PeV cosmic-ray diffusion coefficients reach 10^30 cm^2 s^-1, with perpendicular transport boosted at thermal phase boundaries while overall transport is controlled by trans-Alfvenic war...

  12. Cosmological Analysis with Calibrated Neural Quantile Estimation and Approximate Simulators

    astro-ph.CO 2024-11 unverdicted novelty 6.0

    Calibrated NQE enables unbiased field-level cosmological inference from 2D density maps by training mostly on approximate PM simulations and calibrating with ~100 PP simulations.

  13. Effects of Differential Rotation on the Maximum Mass of Neutron Stars

    astro-ph.HE 2026-07 conditional novelty 5.5

    Differential rotation substantially boosts hyperonic neutron-star maximum mass to the GW190814 range, yet a soft SU(6) EoS still fails the PSR J0740+6620 constraint at 346 Hz.

  14. Planet formation at the inner edge of the dead zone II. Outbursts, rings, vortices, and suppression of planetesimal formation

    astro-ph.EP 2026-06 unverdicted novelty 5.0

    2D radiation-hydrodynamical simulations find accretion outbursts unstable to Rossby-wave instability, forming vortices that suppress planetesimal formation until post-burst quiescence.

  15. Magnetic field dynamics in isolated neutron stars with an external dipole field

    astro-ph.HE 2026-05 unverdicted novelty 5.0

    Long-term numerical-relativity runs find isolated neutron stars with an external dipole relax to a stable mixed poloidal-toroidal field with toroidal energy ≲10% after Tayler saturation.

  16. Magnetic field dynamics in isolated neutron stars with an external dipole field

    astro-ph.HE 2026-05 unverdicted novelty 5.0

    Long-term numerical relativity simulations find that neutron star magnetic fields relax to stable mixed configurations with toroidal energy fraction ≲10% within one Alfvén time after Tayler instability saturation.

  17. foap4: Adaptive mesh refinement with OpenACC, MPI, and p4est

    physics.comp-ph 2026-05 unverdicted novelty 4.0

    foap4 shows that adaptive mesh refinement for Euler gas dynamics runs efficiently on GPUs via OpenACC and MPI even with small 8^3 or 16^3 cell blocks.

  18. Is plasmoid-mediated reconnection really important in accretion flows to drive flares in AGNs?

    astro-ph.HE 2026-02 unverdicted novelty 4.0

    Turbulence-mediated reconnection dominates over plasmoid-mediated reconnection in AGN accretion flows, driving faster reconnection and flares at rates ~0.1 V_A.