Parallel Assembly

Within the solve call, all operators will be assembled according to their configured parameters. When 'parallel = true' is used (not available yet for all DG operators), the operators will be assembled in parallel based on the color partitions within the grid. Hence, the computational grid must provide these partitions, see Documentation of ExtendableGrids.jl on Partitioning for details. Also the sparse system matrix needs to be able to work on different partitions in parallel. Once, the grid has partitions, the solver automatically uses a suitable constructor for the system matrix (MTExtendableSparseMatrixCSC from ExtendableSparse.jl).

Note

DG operators that assemble along cell faces need the option 'edges = true' in the partition call for the grid partitioning, otherwise assembly will be still sequentially.