The Power BI Engine in 20 minutes
summary
Power BI loads data via Power Query M (now ~150 data sources), pushes foldable steps back to the source when possible, and stores imports in the VertiPaq columnstore using dictionary encoding and bit-packing to cut memory dramatically. Import mode (VertiPaq) is recommended over DirectQuery for speed; the formula engine is single-threaded while the storage engine queries segments in parallel, so modelling (star schema, low cardinality, turn off MDX hierarchies) often yields 10x+ improvements and tools like Performance Analyzer and Best Practice Analyzer pinpoint costly measures.
AI-written summary · about
chapters (7)
- 0:00Introduction to the engine
- 1:43Data loading and power query
- 6:18Data storage and compression
- 10:26Modeling best practices
- 13:23Relationship and filter logic
- 16:16Query engine performance
- 19:04Q&A and conclusion
related talks
30mEmpower every BI professional to do more with Fabric
01h 08mAdvanced Modelling and Calculation Using the Power BI Designer
57mBillions of records: the real success story using Microsoft Power BI Embedded
51mAggregations in Power BI
18mEmpowering Business Through Data
51mTroubleshooting Power BI Report Performance
45mAzure DW and Power BI, reporting at the speed of light!
58mPower BI for the Developer