I think "we" should write a #GDAL driver for #Kerchunk
mdim, and classic
how does Kerchunk publish its indexes?
New on the blog: we're discussing our experience using #Kerchunk to improve access times to short-range streamflow predictions generated by NOAA's #NationalWaterModel Predictions Dataset.
In this blog, we discuss our experience using Kerchunk to improve access times to short-range streamflow predictions generated by NOAA’s National Water Model Predictions Dataset, achieving a speedup of 4 times, using 16 times less memory.
In the cloud session, Kerchunk has been mentioned twice as a way to work with legacy formats to optimize them to work in the cloud.
For my graduate #ScientificComputing class I made this little #JupyterNotebook for my students showing how to download #CFS data from #AWS, concatenate with #xarray, and save as a #NetCDF file.
This is the data intake step in an Idaho climate forecast dashboard pipeline the class is working on. It could be automated and sped up significantly. Also need to look at #Kerchunk to see if there's a way to stream only the required data (data is GrIB) https://github.com/LejoFlores/Download_CFS_S3/blob/main/DownloadCFS_files.ipynb