Symplectic Integrator on GPUs

Demonstrating mathematical integrity under GPU parallelization and reduced precision.

View the Project on GitHub OleBo/SymplecticIntegrator

GPU-Accelerated Hamiltonian Integrators

Overview

This project implements GPU-accelerated numerical integrators for simulating nonlinear Hamiltonian systems, with a focus on the Hénon–Heiles system — a classical benchmark for studying chaos and long-term dynamical behavior.

The core idea is to combine:

to simulate millions of trajectories efficiently and accurately.


Why This Project Matters

Simulating physical systems over long time horizons is fundamentally challenging:

This project demonstrates that:

Structure-preserving algorithms + GPU parallelism = scalable and physically correct simulations


Project Documentation

📘 Mathematical Documentation

👉 Symplectic Integrator — Mathematical Documentation

Covers:


🧪 Benchmark Experiment

👉 Benchmark Experiment Documentation

Covers:


⚙️ CUDA Implementation

👉 CUDA Implementation Documentation

Covers:


Key Features


📌 Key Insight

This project highlights a fundamental principle in scientific computing:

Numerical accuracy alone is not enough — preserving the underlying physics is essential.


👤 Author

This project is part of a portfolio exploring: