Section: Subvolume
This section controls subvolume post-processing. Subvolume outputs chunks
of data directly from the computational mesh.
The prefix is the label set in incflo.post_processing. For example
incflo.post_processing = subvol1
- subvol1.type
type: String, mandatory
To use subvolume post-processing, specify with keyword
Subvolume
- subvol1.labels
type: List of strings, mandatory
Similar to the Sampling utility, multiple subvolumes can be defined within a single Subvolume instance, where the top-level label is used to define the fields and output parameters and the bottom-level labels are used to define the type of subvolume and its spatial parameters. Below, the label
chunk1will be used where applicable.
- subvol1.fields
type: List of strings, mandatory
Specify which field arrays should be output within the selected subvolume
- subvol1.int_fields
type: List of strings, optional, default is empty
Specify which integer field arrays should be output within the selected subvolume
- subvol1.derived_fields
type: List of strings, optional, default is empty
Specify which derived field arrays should be output within the selected subvolume
- subvol1.output_rename
type: String, optional
If desired, use a different name for the top-level label when writing. In this example, this would replace the
subvol1label in the naming of the output directories. This option is primarily intended to enable different subvolumes to have the same top-level names despite needing different top-level parameters. In that case, it is the responsibility of the user to ensure the bottom-level names are still different, as this is not checked by the code.
- subvol1.chunk1.type
type: String, optional, default = Rectangular
This specifies the type of subvolume to be used. At the moment, Rectangular is the only available type. Rectangular subvolumes are defined using an origin, number of cells, and cell size.
- subvol1.chunk1.origin
type: Vector<Real>, mandatory
Starting point in three-dimensional space to define the rectangular subvolume. This needs to be the lower left corner of a mesh cell for the subvolume data extraction to work properly.
- subvol1.chunk1.num_points
type: Vector<Int>, mandatory
Number of points (more accurately, cells) in each direction to include in the rectangular subvolume.
- subvol1.chunk1.dx_vec
type: Vector<Real>, mandatory
Cell size, in each direction, for the rectangular subvolume. This is used to determine which mesh level should be used for the subvolume, and, as a result, this input argument must correspond to the resolution of one of the mesh levels in the spatial extent of the subvolume. If a single cell size is specified through the
subvol1.chunk1.dxargument below, then this argument (subvol1.chunk1.dx_vec) is not required.
- subvol1.chunk1.dx
type: Real, optional
Cell size for the rectangular subvolume. This optional argument can be used in place of
subvol1.chunk1.dx_vecto specify a single cell size, representing the cell size in all three directions.
- subvol1.chunk1.chunk_size_vec
type: Vector<Int>, optional
Optional argument for more control over how the data is partitioned within a subvolume. By default, the chunk size will be the max grid size in each spatial direction.