← Library / Lakehouse & Databricks
Slide 1Slide 2Slide 3Slide 4Slide 5Slide 6Slide 7Slide 8Slide 9Slide 10
1 / 10

Use the arrows, swipe, or press ← → to move through the slides.

Lakehouse & DatabricksCarousel · 10 slidesSystem Design Series · Day 17

Slowly Changing Dimensions — Types 1, 2, 3, 4 and 6

When a dimension attribute changes, how you track it decides what history you can query. Each SCD type with its trade-offs in storage, ETL and query complexity.

The trade-off I'd flag

Type 2 is the default for anything regulated. The cost is not storage, it is every downstream query that forgets to filter on the current flag.

← PreviousSnowflake vs Star Schema — when to normalise dimensions Next →Databricks Lakehouse Architecture — the full reference