Quick start

This code:

using Oceananigans

grid = RectilinearGrid(size=(128, 128), x=(0, 2π), y=(0, 2π), topology=(Periodic, Periodic, Flat))
model = NonhydrostaticModel(; grid, advection=WENO())

ϵ(x, y) = 2rand() - 1
set!(model, u=ϵ, v=ϵ)

simulation = Simulation(model; Δt=0.01, stop_iteration=100)
run!(simulation)
[ Info: Initializing simulation...
[ Info:     ... simulation initialization complete (392.482 ms)
[ Info: Executing initial time step...
[ Info:     ... initial time step complete (8.823 seconds).
[ Info: Simulation is stopping after running for 13.995 seconds.
[ Info: Model iteration 100 equals or exceeds stop iteration 100.

runs 100 time steps of a two-dimensional turbulence simulation with 128² finite volume cells and a fifth-order upwinded WENO advection scheme. It's quite similar to the two-dimensional turbulence example.

Visualization

They say that a Makie visualization is worth a thousand Unicode characters, so let's plot vorticity,

using CairoMakie

u, v, w = model.velocities
ζ = Field(∂x(v) - ∂y(u))
compute!(ζ)

heatmap(interior(ζ, :, :, 1))
Example block output

A few more time-steps, and it's starting to get a little diffuse!

simulation.stop_iteration += 400
run!(simulation)

compute!(ζ)
heatmap(interior(ζ, :, :, 1))
Example block output

They always cheat with too-simple "quick" starts

Fine, we'll re-run this code on the GPU then:

using Oceananigans

grid = RectilinearGrid(GPU(), size=(128, 128), x=(0, 2π), y=(0, 2π), topology=(Periodic, Periodic, Flat))
model = NonhydrostaticModel(; grid, advection=WENO())

ϵ(x, y) = 2rand() - 1
set!(model, u=ϵ, v=ϵ)

simulation = Simulation(model; Δt=0.01, stop_iteration=100)
run!(simulation)

Notice the difference? We passed the positional argument GPU() to RectilinearGrid.

Well, that was tantalizing

But you'll need to know a lot more to become a productive, Oceananigans-wielding computational scientist (spherical grids, forcing, boundary conditions, turbulence closures, output writing, actually labeling your axes... 🤯). It'd be best to move on to the one-dimensional diffusion example.