1.3.2 Default Storage Formats

What is the default table storage format in OneLake?

Delta Lake format (Parquet files + Delta transaction log), which provides ACID transactions, time travel, and schema evolution.

What is Delta Lake?

An open-source storage layer that adds ACID transactions, schema enforcement, time travel, and unified batch/streaming to Parquet files.

What is the Delta transaction log?

A deltalog directory containing JSON files that record every change to a Delta table, enabling ACID transactions and time travel.

Can you store non-Delta files in OneLake?

Yes. The Files section of a lakehouse supports any format (CSV, JSON, Parquet, images, etc.). Only the Tables section requires Delta format.

What is V-Order optimization?

A write-time optimization in Fabric that sorts data within Parquet files for faster read performance, especially for analytical queries.

Is V-Order enabled by default in Fabric?

Yes, V-Order is enabled by default when writing to Delta tables in Fabric. It can be disabled in Spark settings if needed.

What file format underlies Delta tables?

Apache Parquet. Delta adds a transaction log layer on top of Parquet files for versioning and ACID compliance.

Can you read Delta tables from outside Fabric?

Yes. Delta is an open format. Tools like Spark, Databricks, or any Delta-compatible reader can access Delta tables via OneLake endpoints.

What is schema enforcement in Delta Lake?

Delta rejects writes that don't match the table's schema. New columns require explicit schema evolution (mergeSchema option) to be added.

What is time travel in Delta Lake?

The ability to query a table as it existed at a previous point in time using VERSION AS OF or TIMESTAMP AS OF syntax.