Persist Data Snapshot for Historical Trend Analysis

[clayhs]

I am attempting to persist calculated data in my Analysis Services Model as snapshots for historical trend analysis in PowerBI. It seems like this my not be doable but am hoping a potential workaround can be found. Ultimatly I am trying to avoid having to export data out of the data model for storage.
Below is a breakdown of what I'm trying to accomplish.
1. Calculated table provides current snapshot of data used in our current day reporting. Calculated values are highlighted.

2. Since calculated values change as the underlying data updates daily, we need to save a snapshot of those values on a regular basis (weekly).
3. This current snapshot data would then be moved to a Historical table, with future weekly snapshot data appended to the now static historical snapshot data.

 

 

 

[v-yiruan-msft]

Hi @clayhs ,

I'm not sure if I understand your requirement... Do you want to copy the "old" daily data into Historical table in Power BI? I'm afraid it can't be achieve in Power BI as Power BI can turn the source data into coherent, visually immersive, and interactive insights.  Which data source these daily data comes from? Whether we can consider to insert these old daily data into a historical table from data source side?

Best Regards

[clayhs]

Thanks for the reply. In a nutshell what I want to do is to take a snapshot of my calulated values so
1. They will no longer change when the underlying data they are calculating from changes
2. I can use those snapshots to see historical trends in how my data is changing.

While I can of course export that snapshot data back into my source database it would be much easier and preferable to somehow store that calculated data in the data model itself (due to other factors of the project I'm working on there would be a number of other moving pieces that would make exporting into the database extra complicated). 
Think something along the lines of incrimental refreash where old data is not refreashed/reprocessed and only new current data is appended to the dataset.