NORI: a GPU-accelerated code for simulating self-gravitating and self-interacting dark matter
Abstract
Self-interacting dark matter (SIDM) can form cores in dark matter haloes and later drive their gravothermal collapse, raising their central densities far above those of NFW haloes. Simulating this collapse is computationally demanding, requiring increasingly short time-steps and a large particle number to resolve the shrinking core. We present NORI, an N-body code that runs on CPUs and GPUs and simulates isolated self-gravitating systems with elastic, isotropic self-interactions using a Monte-Carlo method. To avoid MPI communication in the tree walks, each compute node in CPU mode, and each GPU or group of GPUs in GPU mode, holds a copy of all particles and its own Morton-ordered tree. A fixed evaluation order makes the results bitwise identical on CPUs and GPUs. Applying scattering events sequentially conserves the momentum and energy of each pair to round-off and allows multiple scatterings per particle per step. With global time-steps in both codes, NORI is several times faster on CPUs than our AREPO-2 implementation. One four-A100 GPU node is up to 60 times faster per time-step than AREPO-2 on a 112-core CPU node. For a full core-collapse simulation, where AREPO-2 uses individual time-steps, the speed-up is ∼ 30 for the central density to reach 103--104 times the NFW scale density, ρs. We follow a halo with 108 particles within r200 to a central density of 104 ρs. Within the limit of about 2×109 particles set by data replication, NORI provides an efficient tool for isolated SIDM simulations, including gravothermal collapse and cluster collisions.
Related
- Dynamics of self-gravitating halos with decaying dark matter
- Efficient numerical evolution of self-gravitating systems with dark-matter-baryon scattering
- Strong-lensing perturber signatures in self-interacting dark matter simulations
- Revisiting self-interacting dark matter with DESI DR2 baryon acoustic oscillation observations
- Quantum corrections as a Bound for Detecting Self-Interacting Ultralight Dark Matter
BibTeX
@misc{zier2026nori,
title = {{NORI: a GPU-accelerated code for simulating self-gravitating and self-interacting dark matter}},
author = {Oliver Zier and Ruediger Pakmor and Xuejian Shen and Rongrong Liu and Lucas Schleuss and Mark Vogelsberger and Lars Hernquist},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.37768},
eprint = {2610.37768},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.37768},
}
RIS
TY - UNPB
TI - NORI: a GPU-accelerated code for simulating self-gravitating and self-interacting dark matter
AU - Oliver Zier
AU - Ruediger Pakmor
AU - Xuejian Shen
AU - Rongrong Liu
AU - Lucas Schleuss
AU - Mark Vogelsberger
AU - Lars Hernquist
PY - 2026
DA - 2026/10/27
PB - arXiv
DO - 10.48550/arxiv.2610.37768
UR - https://arxiv.org/abs/2610.37768
AB - Self-interacting dark matter (SIDM) can form cores in dark matter haloes and later drive their gravothermal collapse, raising their central densities far above those of NFW haloes. Simulating this collapse is computationally demanding, requiring increasingly short time-steps and a large particle number to resolve the shrinking core. We present NORI, an N-body code that runs on CPUs and GPUs and simulates isolated self-gravitating systems with elastic, isotropic self-interactions using a Monte-Carlo method. To avoid MPI communication in the tree walks, each compute node in CPU mode, and each GPU or group of GPUs in GPU mode, holds a copy of all particles and its own Morton-ordered tree. A fixed evaluation order makes the results bitwise identical on CPUs and GPUs. Applying scattering events sequentially conserves the momentum and energy of each pair to round-off and allows multiple scatterings per particle per step. With global time-steps in both codes, NORI is several times faster on CPUs than our AREPO-2 implementation. One four-A100 GPU node is up to 60 times faster per time-step than AREPO-2 on a 112-core CPU node. For a full core-collapse simulation, where AREPO-2 uses individual time-steps, the speed-up is ∼ 30 for the central density to reach 10^3--10^4 times the NFW scale density, ρ_s. We follow a halo with 10^8 particles within r_200 to a central density of 10^4 ρ_s. Within the limit of about 2×10^9 particles set by data replication, NORI provides an efficient tool for isolated SIDM simulations, including gravothermal collapse and cluster collisions.
ER -