DP-600 · 3.1.7 · Implement and manage semantic models
Composite models
Design and build composite models
Mix or sources only when one model must combine different freshness, scale, or source requirements.
Mix modes deliberately
A combines or multiple sources in one .
Use DirectQuery when a must remain at the source, Dual for dimensions that frequently interact with DirectQuery facts, and Import for tables that benefit from fast in-memory access.
across are limited and can be slower or less flexible. Start with Import when possible and introduce composite behavior only when a real freshness, scale, or source constraint requires it.
Key points
- Composite: more than one or source
- DirectQuery: fits large or near-real-time fact data
- Dual: improves dimensions beside DirectQuery facts
- Cross-source: become limited
Click an underlined term for a one-line explanation, or see them all in the glossary .
Exam trap
Reaching for a first. gives the best performance and design flexibility, so use it unless data volume or near real-time needs rule it out.
Sources · 3 Microsoft Learn pages Show ▾
- Use composite models in Power BI Desktop
- Composite model guidance in Power BI Desktop
- Model relationships in Power BI Desktop
Checked on October 4, 2026. Report a problem
Go deeper on Microsoft Learn
- Use composite models in Power BI Desktop
- Composite model guidance in Power BI Desktop
- Model relationships in Power BI Desktop
Checked against Microsoft Learn on October 4, 2026. Report a problem with this card