on 09 Nov 2018 20:57:48
Dataflows would be really useful if you could connect to them via DirectQuery. This would mean that we could query enormous dataflows that are not practical to load to the Data model and can make aggregate dataflows that we load to the Data model.
Unsure if this is possible with the architecture of a Dataflow, but it could be quite handy if it is possible.
Administrator on 29 Jun 2021 22:12:43
This has been released: Power BI June 2021 Feature Summary | Microsoft Power BI Blog | Microsoft Power BI
- Comments (37)
RE: Dataflows Direct Query
I agree, this feature is a must have. else we are just doubling our data process and data movement.
RE: Dataflows Direct Query
Another component is to be able to use the analyse in excel feature but against a dataflow instead of a pbi model. Sometimes I just need to pull a single table of a model into a excel file.
RE: Dataflows Direct Query
Yes, a really important feature.
RE: Dataflows Direct Query
and make this working in mix model too, please.
RE: Dataflows Direct Query
Can't imagine a better improvement! I'm currently having to import large amounts of data, making the report sluggish. I can't direct query because we don't have the data warehouse.
RE: Dataflows Direct Query
This would certainly be very useful for us as well. Dataflows are a great way to put data in the cloud. Being able to DQ them would be very useful.
RE: Dataflows Direct Query
The other aspect I am finding the more I tinker with Dataflows is increasing the importance of being able to DQ these for a few reasons. 1) Getting the timing schedules for refreshes has just got a lot more complex - To get a Dataset refreshed you need to have the Dataflow refresh complete to get the latest data, then only after that is finished start refreshing the Dataset to get the latest data. The sequence of timings is going to be complex, plus also time debilitating. My first dataflows are taking around 1 hour to process, and then my datasets after that are taking a bit of time. Waiting a few hours to get the latest data is feeling like a step backwards compared to the refresh times I was getting with my existing models. 2) We can use Dataflows as a way to generate a 'Single Model of Truth' with the data preparation. However, since anyone can modify a DF in their Datasets this also breaks the certainty that the data validity is maintained. By allowing a DQ connection you can be assured that the Single Model of Truth can be maintained.