Try parquet if your data is tabular, pyarrow and related tools are getting parquet up to a pretty comparable speed to hdf5, with arguably more flexibility and a

Try parquet if your data is tabular, pyarrow and related tools are getting parquet up to a pretty comparable speed to hdf5, with arguably more flexibility and a better multithreading story.