OpenFOAM Dam Break: Run in Parallel (decomposeParDict + decomposePar) | Ch7–Part 8

Описание: In Chapter 7 – Part 8, we prepare the OpenFOAM damBreak case to run in parallel. Before you can execute interFoam on multiple CPU cores, you must decompose (split) the mesh and fields into per-processor subdomains.

What we do in this video

✅ Locate and understand system/decomposeParDict
✅ Set the number of processors (example: 4 cores)
✅ Choose the decomposition method: simple

Define splits in (x, y, z)

Rule: nx × ny × nz = number of processors

For this 2D-style case: (2, 2, 1) → 2×2×1 = 4 subdomains

✅ Run the decomposition:

Execute decomposePar (reads system/decomposeParDict automatically)

✅ Verify the output:

New folders created:

processor0, processor1, processor2, processor3

Each processor folder contains:

constant/polyMesh (its local mesh piece)

0/ (its local initial fields)

✅ Read and interpret decomposePar terminal output:

Method used + subdomain count

Cells/points per processor

Neighbor processor connectivity

Load-balance stats (max vs average cell count)

Field transfer (global 0/ → processor*/0/)

At the end of this part, the case is fully decomposed and ready to run interFoam in parallel.

#OpenFOAM #CFD #DamBreak #interFoam #ParallelComputing #decomposePar #decomposeParDict #MPI #MultiphaseFlow #VOF #OpenFOAMTutorial #Linux