Sign up to take part
Registered users can ask their own questions, contribute to discussions, and be part of the Community!
This is not a common pattern. In fact it's an anti-pattern given that Dataiku does have a plugin to export data directly into PowerBI.
PowerBI is not a datastore. At best you could consider it as logical layer on top of other datastores. Microsoft doesn't even provide official ODBC drivers for PowerBI and of course there are no JDBC eithers which is what Dataiku will tipically use. There is however a PowerBI REST API which allows datasets to be queried using anything that can talk to a REST API. In Dataiku Python will be perfect but again Microsoft doesn't provide an official Python package to use the PowerBI REST API so you either do things manually using the Python requests package or you use a third party package like pbipy.
Welcome to the Dataiku Community. We are so glad you have joined.
Another way to think about this is to see if the power BI file is actually the source of the data. In most power BI use cases I've used with Power BI, the Data actually lived somewhere else and it was being pulling into Power BI for manipulation and presentation. Check out your data connections. See if there is actually a more direct source to the data. For example the data may be comming from a SQL database. If this is the case make that same direct connection into Dataiku and bypass Power BI completly. This is likely to produce a more robust solution.
There is a down side to this approach depending on how much "M" you have used to transform the data inside Power BI. If you have done a lot of transformation of the original data source inside Power BI. You might have to re-do those same transformations from the original source to Dataiku. If this is a long term project. I might re-invent my Power BI inside Dataiku, and not use the Power BI at all.
Just an additional point of view.