Storage Format

The volume data provided is stored in the XML Parallel vtkUnstructuredGrid format, e.g, filename.pvtu [‌1]. The Parallel data format does not actually contain any data but references other serial files (Serial vtkUnstructuredGrid) containing the data for each grid partition used for the computation, e.g., filename_partition_number.vtu. Both the .pvtu (or any .vtu) file can be easily read through the open-source Paraview software, a multi-platform application for data analysis and visualization [‌2]. Datasets are delivered as compressed file (.tar.gz).

Instantaneous data format

The dataset consists of a single .pvtu file per snapshot, directly referencing all the nondimensional variables and their spatial derivatives as well as the node positions. For the sake of handling, the instantaneous data was extracted from a limited region of the computational domain shown in Fig. X.

The names of the provided variables are:

pressure, $p$ ;
temperature, $T$ ;
velocity component in x-direction, $u$ ;
velocity component in y-direction, $v$ ;
velocity component in z-direction, $w$ .

The spatial derivatives of each variable are named as variable_di where i={x,y,z}, e.g., u_dy is the derivative of the first component of the velocity vector with respect to the x-direction.

All the quantities are made dimensionless with respect to the reference state.

Statistical data format

The dataset consists of a set of .pvtu files, possibly grouping similar statistics:

averaged_fields.pvtu:
- averaged velocity, ${\overline {u}}_{i}$
- averaged pressure, ${\overline {p}}$
- averaged density, ${\overline {\rho }}$
- averaged temperature, ${\overline {T}}$
averaged_shear_stress.pvtu: ${\overline {\tau _{ij}}}$ (viscous stress)
Reynolds_stress.pvtu: $R_{ij}$