← 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 13

Compaction — Z-Ordering, bin-packing and VACUUM

The Delta Lake maintenance trio: how OPTIMIZE merges small files, how Z-Ordering improves data skipping, and when VACUUM reclaims storage.

The trade-off I'd flag

Run OPTIMIZE on a schedule, not on demand — and never VACUUM with a retention shorter than your longest-running job or time-travel need.

← PreviousHot / Warm / Cold Tiering — lifecycle policies and cost vs latency Next →Snowflake vs Star Schema — when to normalise dimensions