English | 简体中文

Large-scale N-Body particle simulation system supporting million-particle GPU parallel computation and real-time visualization.

• High-Performance GPU Computing — CUDA parallel computation, million-particle real-time simulation
• Multiple Force Algorithms:
  • Direct N² — O(N²) exact computation
  • Barnes-Hut — O(N log N) tree approximation
  • Spatial Hash — O(N) grid-based neighbor search
• Real-Time Visualization — CUDA-OpenGL interop, no CPU-GPU data transfer
• Multi-Integrator — Euler, Velocity Verlet, Leapfrog
• Interactive Control — Camera rotation/zoom, real-time parameter adjustment

• CUDA Toolkit 11.0+, CMake 3.18+, C++17
• OpenGL 3.3+, GLFW, GLEW
• GPU CC 7.0+

mkdir build && cd build
cmake .. -DCMAKE_BUILD_TYPE=Release
cmake --build . -j$(nproc)

# Default simulation
./n_body_sim

# Custom parameters
./n_body_sim --particles 100000 --algorithm barnes-hut --dt 0.001

Tested on RTX 3060

MIT License