Forum Discussion
Data update in report combining Direct Query and OneLake Data Hub (Power BI Semantic Model)
- 1 year ago
I would have expected, that a change to S1 after publishing would immediately take effect on the R2 onlineThat is only true for data, and for varying interpretations of "immediately".
Meta data changes need to be propagated.
Not more comfortable, but less invasive. A prime example is a semantic model that includes an incremental refresh data source. When you make changes to that model in the Desktop and then publish, this will reset (ie destroy) all your incremental refresh partitions and will attempt to refill them. Not only is that wasteful, it may also time out.
So instead you can use ALM Toolkit to propagate only the meta data changes without killing the partition definitions.
Thank you. When you mention Incremental Refresh, is it possible to download via ALM toolkit or Tabular Editor the full report (that has incremental refresh) from the service? Some time ago I was discussing with somebody that it is annoying that you cannot download it even in the template format through regular channels.
- lbendlin1 year agoSuper User
Those tools are for meta data manipulation. They are not data extraction tools.
What you are referring to are the PBIX download limitations. While that list is long, it has gotten shorter recently.
HOWEVER. Incremental Refresh is based on having multiple partitions per table. Power BI Desktop has no concept of that. So even if you were allowed to download a PBIX with incremental refresh data sources you still could not open it in Power BI Desktop.