Partitioning in Tabular SSAS

Partitioning in Analysis Services has, for most devs, generally been a post hoc effort - build your cube first, then during productionisation, add the partitioning schema (or even just let the DBAs do it..).

And this mostly worked, because you would build out the full model, deploy and process, and head out to lunch while the full cube process ran.

In the tabular world, this model simply doesn't work - every time you add a table to the model, the data is queried, and the model is processed. Great.. Now we have to wait for the processing every time.