Glossary

Open table format

A specification that adds transactions, schema management, and versioning to files stored in a data lake.

Also called: OTF, lakehouse table format

An open table format is a specification, usually layered on top of files like Apache Parquet sitting in object storage, that adds the capabilities a database table normally has — atomic transactions, a consistent schema, snapshots, and the ability to time-travel to a prior version — without requiring a proprietary database engine to own the files. Apache Iceberg, Delta Lake, and Apache Hudi are the best-known implementations.

Before open table formats, a data lake built from raw files had no safe way to update or delete individual rows, no concurrency control for multiple writers, and no reliable way to evolve a schema; every change risked partial writes or readers seeing an inconsistent state. Open table formats solve this by keeping a metadata layer that tracks exactly which files make up which version of the table, so writes are atomic and readers always see a consistent snapshot.

They are the technology that makes a data lakehouse possible: object storage as cheap as a data lake, with the transactional guarantees and schema evolution support previously exclusive to a data warehouse. Because the format is open and file-based, multiple engines — a warehouse, a Spark cluster, a query engine — can read and write the same tables without being locked into one vendor, which is the main reason organizations adopt them over a proprietary, warehouse-only format.

Last reviewed September 22, 2026

In the index now

Related terms

Apache Iceberg An open table format adding ACID transactions, schema evolution, and time travel to files stored in a data lake. Apache Parquet An open-source, columnar file format for storing large analytical datasets efficiently in disk or object storage. Data lakehouse A storage architecture that adds warehouse-style structure and transactions directly on top of low-cost data lake storage. Schema evolution The ability of a table to accommodate structural changes, like an added or renamed column, without breaking data. Partitioning Physically splitting a table's data into segments, usually by a column like date, so queries can skip irrelevant ones. Data lake A storage repository that holds raw data of any format cheaply, with structure applied only when the data is read.

Related tools

Apache Iceberg Open-source table format that adds warehouse-style transactions, schema evolution, and time travel to data-lake files. Databricks Managed lakehouse platform, built by Apache Spark's creators, combining Spark-based compute with governance and ML tooling.

Related guides

How to choose a lakehouse platform Choosing a lakehouse means two decisions, not one — which open table format your data sits in, and which platform manages and queries it.