High Performance Direct Gravitational N-body Simulations on Graphics Processing Units -- II: An implementation in CUDA

Abstract

We present the results of gravitational direct $N$-body simulations using the Graphics Processing Unit (GPU) on a commercial NVIDIA GeForce 8800GTX designed for gaming computers. The force evaluation of the $N$-body problem is implemented in ``Compute Unified Device Architecture'' (CUDA) using the GPU to speed-up the calculations. We tested the implementation on three different $N$-body codes: two direct $N$-body integration codes, using the 4th order predictor-corrector Hermite integrator with block time-steps, and one Barnes-Hut treecode, which uses a 2nd order leapfrog integration scheme. The integration of the equations of motions for all codes is performed on the host CPU. We find that for $N>512$ particles the GPU outperforms the GRAPE-6Af, if some softening in the force calculation is accepted. Without softening and for very small integration time steps the GRAPE still outperforms the GPU. We conclude that modern GPUs offer an attractive alternative to GRAPE-6Af special purpose hardware. Using the same time-step criterion, the total energy of the $N$-body system was conserved better than to one in $10^6$ on the GPU, only about an order of magnitude worse than obtained with GRAPE-6Af. For $N \apgt 10^5$ the 8800GTX outperforms the host CPU by a factor of about 100 and runs at about the same speed as the GRAPE-6Af.

Accepted for publication in New Astronomy

References in corpus (12)

• Star cluster ecology IVa: Dissection of an open star cluster---photometry
• Harvesting graphics power for MD simulations
• GRAPE-6A: A single-card GRAPE-6 for parallel PC-GRAPE cluster system
• High Performance Direct Gravitational N-body Simulations on Graphics Processing Units
• High Performance Direct Gravitational N-body Simulations on Graphics Processing Unit I: An implementation in Cg
• Performance Tuning of N-Body Codes on Modern Microprocessors: I. Direct Integration with a Hermite Scheme on x86_64 Architecture
• Graphic-Card Cluster for Astrophysics (GraCCA) -- Performance Tests
• N-Body Simulations on GPUs
• The Chamomile Scheme: An Optimized Algorithm for N-body simulations on Programmable Graphics Processing Units
• Performance analysis of direct N-body algorithms for astrophysical simulations on distributed systems
• Direct Simulation of Dense Stellar Systems with GRAPE-6
• High-Performance Small-Scale Simulation of Star Clusters Evolution on Cray XD1

Cited by in corpus (57)

• Data Mining and Machine Learning in Astronomy
• Accelerating NBODY6 with Graphics Processing Units
• NBODY6++GPU: Ready for the gravitational million-body problem
• A multiphysics and multiscale software environment for modeling astrophysical systems
• A sparse octree gravitational N-body code that runs entirely on the GPU processor
• Harvesting graphics power for MD simulations
• GAMER: a GPU-Accelerated Adaptive Mesh Refinement Code for Astrophysics
• SAPPORO: A way to turn your graphics cards into a GRAPE-6
• The GENGA Code: Gravitational Encounters in N-body simulations with GPU Acceleration
• The Ecological Impact of High-performance Computing in Astrophysics
• Accelerated molecular dynamics force evaluation on graphics processing units for thermal conductivity calculations
• 24.77 Pflops on a Gravitational Tree-Code to Simulate the Milky Way Galaxy with 18600 GPUs
• On the onset of runaway stellar collisions in dense star clusters - II. Hydrodynamics of three-body interactions
• Survivability of planetary systems in young and dense star clusters
• Teraflop per second gravitational lensing ray-shooting using graphics processing units
• Air pollution modelling using a graphics processing unit with CUDA
• Fast Calculation of the Lomb-Scargle Periodogram Using Graphics Processing Units
• Computational advances in gravitational microlensing: a comparison of CPU, GPU, and parallel, large data codes
• Graphic-Card Cluster for Astrophysics (GraCCA) -- Performance Tests
• Gravitational tree-code on graphics processing units: implementation in CUDA
• Hierarchical N-body simulations with auto-tuning for heterogeneous systems
• Accelerating Dust Temperature Calculations with Graphics Processing Units
• NBSymple, a double parallel, symplectic N-body code running on Graphic Processing Units
• Implementation of a Parallel Tree Method on a GPU
• Swarm-NG: a CUDA Library for Parallel n-body Integrations with focus on Simulations of Planetary Systems
• Parallel Algorithm for Solving Kepler's Equation on Graphics Processing Units: Application to Analysis of Doppler Exoplanet Searches
• Simulation of reaction-diffusion processes in three dimensions using CUDA
• The moment of core collapse in star clusters with a mass function
• Kinematic Modelling of Disc Galaxies using Graphics Processing Units
• Astrophysical Supercomputing with GPUs: Critical Decisions for Early Adopters
• CUDAEASY - a GPU Accelerated Cosmological Lattice Program
• Analysing Astronomy Algorithms for GPUs and Beyond
• PyCOOL - a Cosmological Object-Oriented Lattice code written in Python
• Dust-Dust Collisional Charging and Lightning in Protoplanetary Discs
• Computational Physics on Graphics Processing Units
• Parallel Astronomical Data Processing with Python: Recipes for multicore machines
• N-body simulation of a clumpy torus: application to active galactic nuclei
• A parallel gravitational N-body kernel
• MYRIAD: A new N-body code for simulations of Star Clusters
• Large-scale Ferrofluid Simulations on Graphics Processing Units
• A connected component-based method for efficiently integrating multiscale $N$-body systems
• Simulations of the Solar System's Early Dynamical Evolution with a Self-Gravitating Planetesimal Disk
• How to obtain efficient GPU kernels: an illustration using FMM & FGT algorithms
• Near-Field Effects of Cherenkov Radiation Induced by Ultra High Energy Cosmic Neutrinos
• Finite element modelling of perturbed stellar systems
• Application of Graphics Processing Units to Search Pipeline for Gravitational Waves from Coalescing Binaries of Compact Objects
• A pilgrimage to gravity on GPUs
• Distributed N-body Simulation on the Grid Using Dedicated Hardware
• New features of parallel implementation of N-body problems on GPU
• GraviDy, a GPU modular, parallel direct-summation $N-$body integrator: Dynamics with softening
• Fast Simulations of Gravitational Many-body Problem on RV770 GPU
• Oct-tree Method on GPU
• Gravitational Dynamics of Large Stellar Systems
• Efficient magnetohydrodynamic simulations on graphics processing units with CUDA
• Dynamical Processes in Globular Clusters
• Direct Calculation of Self-Gravitational Force for Infinitesimally Thin Gaseous Disks Using Adaptive Mesh Refinement
• Sapporo2: A versatile direct $N$-body library