Databricks buys Tabular and unifies data lakehouse standards

Databricks buys Tabular and unifies data lakehouse standards

Databricks recently acquired the data management company Tabular. According to the company, this acquisition should ensure the creation of a common standard for data lakehouses.

Databricks’ acquisition of Tabular combines two format standards for data lakehouse environments. No financial disclosures were made about the acquisition, but The Wallstreet Journal estimates the final amount to be more than $1 billion (over 920 million euros). Tabular, incidentally, was also in the interest of Snowflake and Confluent. The acquisition will allow Databricks to further compete with these two data and analytics giants.

Read more: Databricks moves from lakehouse to data intelligence

Delta Lake UniForm and Iceberg

Data lakehouses are architectures for storing large amounts of raw data. Thereby providing structure and management functionality through certain formats.

Databricks offers its Delta Lake architecture, more specifically, its Delta Lake UniForm format, for this purpose, while Tabular has its own format, Iceberg. Iceberg is a high-performance open-source database format that optimizes tables for big data, ensuring that data engines can work with these tables.

Both Delta Lake UniForm and Iceberg operate based on the Apache Parquet data storage format, yet the two are not interoperable. According to Databricks and Tabular, the acquisition should make this happen. Both companies state that the new standard will be developed in close cooperation with Tabular’s existing Apache Iceberg community.

Customers can use the best features of both solutions with the new format and no longer have to worry about which format to use. For Databricks, this allows its own platform to gain more traction with large enterprise customers. This is because fewer competing data lakehouse formats remain.