Snowflake Inc. today said it has extended its core capabilities — including secure data sharing, security and performance optimization — to Apache Iceberg, the fast-growing open-source table format.

The company said the integration allows organizations to employ Iceberg tables without data migration, enabling analytics and artificial intelligence development directly within open-format environments.

Organizations can now run artificial intelligence and analytical workloads directly on open-format data to simplify management and reduce costs. Snowflake said its implementation provides combines the flexibility of open-source interoperability with its security and governance features to effectively remove traditional barriers to open-source adoption.

Snowflake and rival Databricks Inc. have been locked in a competition over who is more committed to supporting Iceberg, which competes with both providers’ own table formats. With this release, “Customers can work with their open data exactly as they would with data stored in the Snowflake platform,” said Christian Kleinerman, executive vice president of product at Snowflake.

That includes data lakehouse analytics, the full range of Snowflake security features and data-sharing capabilities. Snowflake said it’s extending its data replication and synchronization capabilities to Iceberg tables in a feature available in private preview. That enables customers to quickly restore data without major disruptions in the event of a failure, cyberattack or disasters.

Apache Iceberg has rapidly gained traction thanks to its flexible management of large-scale data lakes, schema evolution and transactional capabilities. Snowflake said its Iceberg support is part of a broader commitment to open-source software, as demonstrated by the 35% of its acquisitions over the past four years that have supported open-source technologies.

The company said it contributes to multiple open-source initiatives, including Iceberg, Apache NiFi, the Apache Polaris catalog it created and contributed to open source, pandas acceleration, Apache Streamlit and TruEra large language model observability software.

Image: theCUBE Research/DALL-E 3