Table Engines
Databend provides several table engines so that you can balance performance and interoperability needs without moving data. Each engine is optimized for a specific scenario—ranging from Databend’s native Fuse storage to external data lake formats.
Available Engines
| Engine | Best For | Highlights |
|---|---|---|
| Fuse Engine Tables | Native Databend tables | Snapshot-based storage, automatic clustering, change tracking |
| Apache Iceberg™ Tables | Lakehouse catalogs | Time-travel, schema evolution, REST/Hive/Storage catalogs |
| Apache Hive Tables | Hive metastore data | Query Hive-managed data stores through external tables |
| Delta Lake Engine | Delta Lake datasets | Read Delta tables in object storage with ACID guarantees |
Choosing an Engine
- Use Fuse when you manage data directly inside Databend and want the best storage and query performance.
- Choose Iceberg when you already manage datasets through Iceberg catalogs and need tight lakehouse integration.
- Configure Hive when you rely on an existing Hive Metastore but want Databend’s query engine.
- Select Delta to analyse Delta Lake tables in place without ingesting them into Fuse.