Skip to content

Latest commit

 

History

History
41 lines (31 loc) · 1.31 KB

README.md

File metadata and controls

41 lines (31 loc) · 1.31 KB

3D_Interpolator_Python

Jupyter notebook demonstrating use of Dask and Xarray for lightning fast 3-D interpolation of DNS data.
Dask provides parallelism and array chunking, while Xarray provides convienient writing and compression of NETCDF4 files.
Spline interpolation function is translated from IDL by Swike from StakedOverflow

Prerequisites:
Python

  • Tested on Python 3.11.x Jupyter Notebook 
pip install jupyter notebook

Dask 

pip install dask[complete]

Xarray 

pip install xarray

NETCDF4-python 

pip install netcdf4

Please adapt the code to your own needs.

Due to my needs specifically, the notebook reads coordinate data from:

  • course mesh:     ('geometry_old.nc')
  • fine mesh:      ('geometry.nc')
  • course flow field data: ('flowXXXX.nc')

and attempts to interpolate data (for instance the temperature field 'temp') from the coarse grid to the fine grid.

Arrays are loaded using the Dask.array.from__array function to ensure large arrays do not overrun available RAM on system.
Final flowfile is written using Xarray with zlib compression (specified in the encoding dict)