We are excited to announce new improvements and enhancements to Azure Data Lake Storage Gen2 (ADLS Gen2) support inside Dataflows in Power BI. This includes support for workspace admins to bring their own ADLS Gen2 account, improvements to the Dataflows connector in Power BI, take ownership support for dataflows using ADLS Gen2 and also minor improvements to detaching from ADLS Gen2.
These changes will begin rolling out in the week of August 10th, and can take a few days to reach everyone.
Workspace Admins can now bring their own data lake
One of the biggest pain points we heard from you was that moving all your organizational data to a single ADLS Gen2 account was too restrictive. In this month’s update we have removed this limitation and now allow workspace admins to connect to an Azure subscription to bring their own ADLS Gen2 account. This will make it easier for departments to control and assign permissions as well as give flexibility to large organizations who may require multiple ADLS Gen2 accounts for different needs and purposes.
To use this feature, first head toward a workspace which has no dataflows (Note: you cannot connect to an ADLS Gen2 account if there are dataflows defined in that workspace). Click on Workspace settings and you will see a new tab called Azure Connections. Click on this tab and click the Storage section.
Click ‘Connect to Azure’ and Power BI will retrieve a list of Azure subscriptions you have access to. Fill in the dropdowns and select a valid Azure subscription, resource group and storage account which has hierarchical namespace option enabled (this is the ADLS Gen2 flag).
Once selected, click Save and you now have successfully attached the workspace to your own ADLS Gen2 account. You will notice that Power BI will automatically configure the storage account with the required permissions and set up the PBI filesystem where the data will be written to. Now every dataflow’s data inside this workspace will write directly to this filesystem. This can then be used with other Azure services creating a single source for all of your organizational or departmental data.
Other improvements to ADLS Gen2 for dataflows
As part of this update we have made some small improvements to other features related with ADLS Gen2 inside Power BI including:
- Workspace users can now take ownership of dataflows which are using ADLS Gen2 storage. Prior to this update, take ownership only worked if you were using the Power BI managed lake. This limitation has now been removed.
- The Dataflows connector now allows any user inside a workspace connect to the dataflow. Prior to this update, if you were using ADLS Gen2, only the owner of the dataflow could connect to the dataflow inside Power BI Desktop. This limitation has now been removed.
- Full support for detaching or removing the tenant or workspace from an ADLS Gen2 account. This is very helpful if you wish to first try out the ADLS Gen2 account or if you removed it and no longer want to associate your tenant or workspace to an ADLS Gen2 account. Prior to this update, if you needed to detach your ADLS Gen2 account from the tenant, you needed to first raise a support ticket with us. This limitation has now been removed. Note to detach from a tenant, all workspaces must be detached first. To detach an ADLS Gen2 account from a workspace, all dataflows inside that workspace must be deleted before we allow the detach to continue.
- This feature is supported for Power BI Pro, Premium and Embed (all the new APIs are public and can be used to automate this process)
- This feature will be available in all regions
- The ADLS Gen2 account must be in the same region as your Power BI tenant and the resource must belong to a subscription inside the same tenant.
- The user connecting to ADLS Gen2 must be in the Owner role of the subscription or resource group or resource. If you are the subscription admin, you will still need to add yourself as owner to one of these levels before continuing. This requirement will be patched in subsequent releases Multi-Geo ADLS Gen2 accounts are currently not supported
That’s all for this month! We hope that you enjoy these updates for the month. If you have any feedback feel free to contact the dataflows team