I am curious if anyone has already tried the following things.
With XMLA end point read-write enabled, used SSMS to refresh the datasets or to override the incremental refresh ? Reason I am interested: It takes away the limitations PBI service has on 48 refreshes per day and no refresh timeout is imposed.
Used custom queries to detect data changes during incremental refresh? Reason I am interested: I read custom query is lot more efficient as it uses lightweight M expressions.
Used ALM tool kit to deploy metadata only changes to the dataset ? Reason I am interested: This is important specially for data sets in premium capacity with incremental refresh enabled. If I need to update the data model with couple of new columns/measures and with current limitation of NO DOWNLOAD of pbix file for incremental refresh enabled data sets, this is a life saver.
So I will really appreciate if someone can point me in the right directions. I have found some MS documents on these but dont provide step-step by instructions on how to successfully implement.
This document is a starter and sort of defines everything I listed above but does not go in detail in setting things up.