CliMA/Oceananigans.jl

🌊 Julia software for fast, friendly, flexible, ocean-flavored fluid dynamics on CPUs and GPUs

What it solves

Oceananigans.jl provides a high-performance framework for simulating incompressible fluid dynamics, specifically tailored for ocean modeling. It addresses the need for a flexible, GPU-accelerated simulation tool that can handle both nonhydrostatic and hydrostatic Boussinesq equations across different domains (Cartesian and spherical shell).

How it works

The software uses finite volume simulations to solve dynamical equations. It is written in Julia and designed to be performance-portable, allowing users to switch between CPU and GPU execution by simply changing a single keyword (e.g., CPU() to GPU()). It supports various advection schemes (like WENO) and can be integrated with other tools for differentiability (via Enzyme.jl and Reactant.jl).

Who it’s for

It is designed for researchers and developers in academia and industry who perform high-resolution ocean modeling, geophysical fluid dynamics, and climate simulations.

Highlights

  • GPU Acceleration: Optimized for Nvidia GPUs, enabling high-resolution global simulations with O(10 km) grid spacing.
  • Flexible Interface: A user-friendly API that simplifies the setup of both simple and complex simulations.
  • Performance Portability: Seamless switching between CPU and GPU backends.
  • Extensible: Supports ML-augmented parameterizations and differentiable workflows for Earth system modeling.

Related

  • Project
    omlins/ParallelStencil.jl

    A Julia library for architecture-agnostic, high-performance parallel stencil computations that allows domain scientists to write high-level code deployable on both CPUs and GPUs.

  • Project
    google/jax-cfd

    JAX‑CFD is a research‑grade, JAX‑based library for differentiable CFD. It provides finite‑volume and pseudospectral solvers, optional ML‑augmented models, and data utilities, enabling gradient‑based experiments on turbulent flow simulations. The package is installable via pip, comes with Colab demos, and is now unmaintained, with newer alternatives recommended.

  • Project
    SciML/DifferentialEquations.jl

    A high-performance numerical solver suite for a wide variety of differential equations, supporting scientific machine learning and GPU acceleration.

  • Project
    dynamiqs/dynamiqs

    A GPU-accelerated and differentiable Python library for high-performance simulation of quantum systems, designed for large-scale problems and gradient-based parameter estimation.

  • Project
    SciML/DiffEqGPU.jl

    A GPU acceleration library for differential equation solvers that enables parallel solving across many parameters and initial conditions.