Python running out of memory when writing a raster using Rasterio

Question

I am working on a script in which I am rasterising a vector using Rasterio.

When I try to write the raster to a tif file, my cluster runs out of memory and crashes about 30 seconds after the job starts.

Is there a more memory friendly way in which I can save this as a tif file using Rasterio?

My code is below:

with rasterio.open(
        os.path.join(dir, filename), "w",
        driver="GTiff",
        transform=raster.transform,
        dtype=rasterio.float32,
        count=1,
        crs=raster.crs,
        width=raster.width,
        height=raster.height) as dst:
    dst.write(array, indexes = 1)

Answer 1

You could use rioxarray (library that mixes rasterio and xarray) to chunk your input data and write it using dask.