PBIMCR - Direct Query with Chris Webb

In DirectQuery mode Power BI does not store any data inside a dataset but instead runs queries against your data source to get the data in needs when a report is rendered. It's very powerful - but you can easily run into performance issues if you do things the wrong way. In this session you'll learn the best practices for using DirectQuery mode in Power BI: how data should be modelled, which properties and settings are important, and when you should use composite models and aggregations.
Chris Webb is a ..
specialist in Power BI, Analysis Services, MDX, Power Pivot, DAX, and Power Query. He's been blogging about Microsoft BI for over 10 years, and racked up over 1500 posts.! He also speaks at a number of conferences including SQLBits, the PASS Summit, the PASS BA Conference, SQL Saturdays and user groups.

[@Bharath_PBI]

Awesome 👍 Looking for more from Chris Webb, his blog is just amazing. Thank you!!

[@yugen4070]

Hope we see a direct query performance optimization techniques video

[@charlesdelanoy]

This is great wondering if there is similar for import mode ?

[@BenHoward_PowerBI]

You mean "Best Practices for Import Mode"?

[@laionegan]

Very interesting. Thx

[@asadm540]

What is 'really really large' amounts of data - cannot be measured in row count (2 bn.) . Also - most grown up data models can't be flattened down to a single table. So, you'll end up importing 10-12 tables into import mode, replicating your data model and your data. Get your users hooked on import mode and your data grows and vertipaq and import mode chokes and you have to re-engineer. Paginated reports (ex SSRS) does not handle sharing the same imported dataset for 100s of concurrent users.