On the Power BI team we want our customers to be successful whichever data source they’re using – not just the Microsoft ones. Recently I had the pleasure of recording a webinar on the subject of best practices for using Power BI with Snowflake, along with Craig Collier from Snowflake and Chris Holliday from Visual BI (a partner that specialises in Power BI/Snowflake solutions). You can watch it here:
https://info.microsoft.com/ww-landing-build-scalable-BI-solutions-using-power-BI-and-snowflake.html
There’s a summary of what we talk about on the Snowflake blog:
https://www.snowflake.com/blog/maximizing-power-bi-with-snowflake/
We’re continuously improving the performance and functionality of all our connectors, and Snowflake is no exception: we recently announced that in early 2021 we’ll have support for Snowflake roles and the ability to use your own SQL queries in DirectQuery mode. Power BI support for Azure Service Tags (on the roadmap here) will also be very useful for Snowflake customers.
Great news – but why native queries available only in DirectQuery? Why not import mode?
And will this mean that query folding works as well? For incremental refresh purposes?
Hi Chris. This is timely. Question: Do you have any horror stories about what kinds of needless compute costs that clients were paying by not following the kinds of best practices you cover in this video? And also by relying on blanket DirectQuery over Snowflake instead of taking a managed approach that included aggregations, composite models, and the like?