is it possible to write chinese and other internal data.bases in Fortran so that it is faster?

I was thinking using data.table.

I think I can try both approaches with different new functions until replacing the original. E.g: