Adaptive Mesh Fluid Simulations on GPU
arXiv:0910.5547 · doi:10.1016/j.newast.2009.10.002
Abstract
We describe an implementation of compressible inviscid fluid solvers with block-structured adaptive mesh refinement on Graphics Processing Units using NVIDIA's CUDA. We show that a class of high resolution shock capturing schemes can be mapped naturally on this architecture. Using the method of lines approach with the second order total variation diminishing Runge-Kutta time integration scheme, piecewise linear reconstruction, and a Harten-Lax-van Leer Riemann solver, we achieve an overall speedup of approximately 10 times faster execution on one graphics card as compared to a single core on the host computer. We attain this speedup in uniform grid runs as well as in problems with deep AMR hierarchies. Our framework can readily be applied to more general systems of conservation laws and extended to higher order shock capturing schemes. This is shown directly by an implementation of a magneto-hydrodynamic solver and comparing its performance to the pure hydrodynamic case. Finally, we also combined our CUDA parallel scheme with MPI to make the code run on GPU clusters. Close to ideal speedup is observed on up to four GPUs.
arXiv commentSubmitted to New Astronomy
References in corpus (8)
- PLUTO: a Numerical Code for Computational Astrophysics
- The Formation of the First Star in the Universe
- Athena: A New Code for Astrophysical MHD
- Dissipation in Compressible MHD Turbulence
- Kinetic Energy Decay Rates of Supersonic and Super-Alfvenic Turbulence in Star-Forming Clouds
- Magnetohydrodynamic Simulations of Disk Galaxy Formation: the Magnetization of The Cold and Warm Medium
- Relativistic Hydrodynamic Flows Using Spatial and Temporal Adaptive Structured Mesh Refinement
- Nonlinear Evolution of Gravitational Fragmentation Regulated by Magnetic Fields and Ambipolar Diffusion
Cited by in corpus (26)
- Enzo: An Adaptive Mesh Refinement Code for Astrophysics
- Magnetic Fields in Population III Star Formation
- Resolved magnetic dynamo action in the simulated intracluster medium
- On the amplification of magnetic fields in cosmic filaments and galaxy clusters
- Simulations of extragalactic magnetic fields and of their observables
- Cholla : A New Massively-Parallel Hydrodynamics Code For Astrophysical Simulation
- GAMER-2: a GPU-accelerated adaptive mesh refinement code -- accuracy, performance, and scalability
- The small scale dynamo and the amplification of magnetic fields in massive primordial haloes
- Simulations of ultra-high Energy Cosmic Rays in the local Universe and the origin of Cosmic Magnetic Fields
- Dynamical evolution of magnetic fields in the intracluster medium
- Forecasts for the detection of the magnetised cosmic web from cosmological simulations
- Magnetic fields during the formation of supermassive black holes
- A survey of the thermal and non-thermal properties of cosmic filaments
- Detecting shocked intergalactic gas with X-ray and radio observations
- Enhancing the Spectral Hardening of Cosmic TeV Photons by Mixing with Axionlike Particles in the Magnetized Cosmic Web
- Evolution of primordial magnetic fields during large-scale structure formation
- PyCOOL - a Cosmological Object-Oriented Lattice code written in Python
- Multi wavelength cross-correlation analysis of the simulated cosmic web
- The evolution of cosmic ray electrons in the cosmic web: seeding by AGN, star formation and shocks
- Role of magnetic fields in the formation of direct collapse black holes
- Inflationary and phase-transitional primordial magnetic fields in galaxy clusters
- GPU-Native Adaptive Mesh Refinement with Application to Lattice Boltzmann Simulations
- Numerical cosmology on the GPU with Enzo and Ramses
- Load Balancing For High Performance Computing Using Quantum Annealing
- Efficient Tsunami Modeling on Adaptive Grids with Graphics Processing Units (GPUs)
- Hardware-Accelerated Phase-Averaging for Cavitating Bubbly Flows