Power BI Gateway still required if migrating on premise datasources to Azure tenany

[Anonymous]

Hi,

I currently have a number of 'on-premise' servers that we pull data from, we also have a Power BI Gateway that is configured to communicate to a Power BI Premium SaaS, we are looking to migrate the on-premise application servers into our new Azure tenancy, so all servers will still look and feel like they do today on premise but just deployed to Azure.

My question is, would I still require the Power BI Gateway in front of these application servers now that they are in Azure?

As a side point, the Power BI Premium SaaS we have a license for is in a partner company tenancy, all of our users use our partners tennancy ccounts to logon to the Power Bi Service and expect this to still continue as normal.

Kind regards,

[v-cazheng-msft · Accepted solution]

Hi @Anonymous ,

Normally, we don’t need configure on-premises data gateway for cloud data sources for it can be directly accessed by Power BI. But if Power BI couldn’t straightly connect to the data source, on-premises data gateway is still needed even it’s a cloud data source. For most data sources in Azure, gateway is not required since they are in the same private network. But there does exist some exceptions. Here is a list contains data sources supported by Power BI and whether gateway is needed for them. You may refer to it and make relevant configurations.

Power BI data sources - Power BI | Microsoft Docs

If there is any post helps, then please consider Accept it as the solution to help the other members find it more quickly. If I misunderstand your needs or you still have problems on it, please feel free to let me know. Thanks a lot!

Best Regards,

Community Support Team _ Caiyun

[Tutu_in_YYC]

Theoretically you dont need Power BI Gateway anymore if the source is in the cloud. But you can still use it as "credential repo". This is great if you have multiple report/datasets using the same data source, and you can configure it in in Gateway, instead of in every dataset. It makes managing the data source credentials easier.