Data Tools disabled with PowerBi as Service

[mrvamsidhar]

Hello, Up until noe I download data to my powerbi desktop using Azure enterprise connetors and prepare necessary reports. I used to create a custom tables on top of my data and used to publish it. Now that, Our team is increased and people are working on different kind of reports on the same data set I decided to move to Power-Bi as service.

I created a shared workspace and created a dataset in app.powerbi. Now that my data set is created, I was able to connect to this data set as a service from my Power-Bi Deskttop. 

Now I have to create different custom tables on top of  my data. But i see all the modelling options are disabled. (becase it is connected to live data set). 

How can I create Custom tables on the data set conncted as a service?  Am I doing some thing wrong? 

Also, FYI -  I am using powerbi Desktop version from October 2017. Is latest version fix this problem?

I cannot use latest Desktop version because, Azure enterprise connector is removed from latest version and the new Azure Consumption insights connector which was introduced is slow in downloading months worth of data. I already reported this)

[Seth_C_Bauer]

@mrvamsidhar You are correct, the connection is a live connection and is treating the dataset in the Service as the source (Analysis Services Tabular Model) - which it is. So, you will need to update the original report with the new tables added to the model to support your efforts in this other report.

[mrvamsidhar]

@Seth_C_Bauer,Thanks for your reply. I am loading my Data set with Azure Enterprise Data Connector. I don't have control to update the data set.

With the enterprise data I have to map with my Business data. I am not sure how to do this. Is there any way I can  connect to Live to live only when needed? So that, I can put all my custome tables and refresh data when needed.

[Seth_C_Bauer]

@mrvamsidhar As of today, a Live & Import scenario is not supported. Your only option would be to connect to the AS instance and import from there to join with a different dataset. This most likely may not be feasible due to data size, but that would be the only option right now.

[mrvamsidhar]

Thanks again. Thats pretty basic requirement. Isn't? Now that we have a team and every one is working on creating different reports individually on the dataset it is craeating a lot of confusion on sharing these...