Data modeling for experts with Power BI
summary
Covers practical Power BI behavior: how table-level storage modes (Import, DirectQuery, Dual) create composite models and separate source groups, how cross–source-group joins become limited relationships (they act as inner joins so the “blank” unmatched bucket disappears), and the performance/security tradeoffs when unique keys flow between sources including the empirical rule of keeping unique one-side keys under 50,000 for acceptable DirectQuery latency. Shows using external tools to author calculation groups, explains remote vs local measures (remote DAX is hidden) and when calc groups won’t apply because a measure isn’t “wholesalable.”
AI-written summary · about
chapters (9)
- 0:00Welcome and brain melt edition
- 4:57Storage modes in Power BI
- 9:32Composite models introduction
- 15:32Demo composite models
- 24:43Composite models on Power BI datasets
- 37:07Calculation groups overview
- 41:17Wholesale vs non-wholesale measures
- 45:28Demo calculation group behavior
- 50:44Summary and resources
related talks
16mGood Data Modeling & Effective Communication in Power BI
01h 08mAdvanced Modelling and Calculation Using the Power BI Designer
51mAggregations in Power BI
54mData Visualization with Power View and the Tabular Model
50mLessons learned: Governance and Adoption for Power BI
51mTroubleshooting Power BI Report Performance
21mUnlocking Synergy: PowerBI and PowerApps Collaboration for Data-Driven Decision-Making
48mTime Intelligence in Power BI