Tutorial

Installation

In your personal workstation, the installation of KitAMR.jl is quite simple. Just run

julia> ]
(//your_julia_version//) pkg> activate .
(//your_path//) pkg> add KitAMR

For cluster environment, preliminary procedure is required to deploy a local p4est, which enable KitAMR.jl to use the optimized MPI backend on cluster.

Utilization

To launch a simulation, user is required to execute a Julia script from the command line in an MPI manner. Examples can be found in /example. Here we show one of them and provide detailed explanation.

X-38 spacecraft

In KitAMR.jl, user-provided configuration can be input in two approaches.

The first one is to define a .txt file, that is /example/X38/configure_X38.txt. The file will be read by read_config and returned as a dictionary. In the file, names in front of = are required to be consistent with the fieldnames of Configure type and the types in its field (including Gas, Solver, Output, and UDF). The meaning of these fields can be referred to the definition of these types in Type pages.

The other one is to construct the Configure directly in .jl script.

The scripts in these two approaches are respectively given by /example/X38/X38_text.jl and /example/X38/X38_julia.jl.

The initial conditions function in PCoordFn and other user-defined functions in UDF are collected in /example/X38/X38_udf.jl.

The geometry of the immersed boundary is defined in /example/X38/X38_normalized.stl.

For other information, one may refer to the comments in /example/X38/X38_julia.jl.