Forum Discussion
Take Data Stream as table for Data Modeling
Hi,
It seems that I can only take one Data Stream API (historic data analysis-enabled) as one table in one dataset.
But I would like the data table produced by the Data Stream as part of my data model, so I can write some DAX to explore it.
And it can not be done. Not by Data Stream, at least. Because the Data Stream it self is already a dataset, I assume.
Do I have to change the Data Stream into another style of data feeding to get it done? Another style which is "static" data can be put into Data Model?
Thanks!
Hi. Let's see if I understood to help. Streaming datasets are composed by one table. Once it's created you can consider it as a Power Bi Dataset that can be conected with Power Bi Desktop as live conection (having the ability to create measures only).
If you want to merge that data with some other data you have somewhere else, you should consider getting the data from the source that is pushing to the streaming and not getting it from the streaming data set that you have just generated. If you don't have a data source because you are pushing something that is created on the fly, consider adding to the script that is pushing the data to save it on a database too and not only sending it on power bi. That way you can keep track of the historic streaming pushes on a database to mix with other data.
I hope it makes sense.
1 Reply
- ibarrauSuper User
Hi. Let's see if I understood to help. Streaming datasets are composed by one table. Once it's created you can consider it as a Power Bi Dataset that can be conected with Power Bi Desktop as live conection (having the ability to create measures only).
If you want to merge that data with some other data you have somewhere else, you should consider getting the data from the source that is pushing to the streaming and not getting it from the streaming data set that you have just generated. If you don't have a data source because you are pushing something that is created on the fly, consider adding to the script that is pushing the data to save it on a database too and not only sending it on power bi. That way you can keep track of the historic streaming pushes on a database to mix with other data.
I hope it makes sense.