We have similar (above ground, but still labour intensive) aqueducts on the dry side of the Alps, and you can see early beginnings of data processing in the sol

We have similar (above ground, but still labour intensive) aqueducts on the dry side of the Alps, and you can see early beginnings of data processing in the solutions they came up with for the problem of "we want people to take water out of the system in direct proportion to how much labour they contribute to its maintenance" — including medieval one-way functions like broken tallies.