Snowflake is launching a PostgreSQL database-as-a-service within its AI data environment to place transactional workloads alongside analytics and AI under a single set of governance rules.

PostgreSQL now runs natively in the AI Data Cloud, allowing teams to power apps and AI agents, analyze business performance and trends using the data from their operations, and build recommendations or forecasting systems without the cost and complexity of building data pipelines and managing multiple vendors, the cloud data platform provider said.

“Say you want to build an app on data that is in Snowflake, but if that app doesn't have a relational OLTP [online transaction processing] database to store the data, they have to go break out of the boundary," Snowflake EVP of product Christian Kleinerman told The Register . "With the PostgreSQL service, our goal is to provide this secure boundary where, if customers build apps or build agents within that boundary, their data has not left the compliance and regulatory perimeter for Snowflake.”

[1]

In addition, the service's full compatibility with open source PostgreSQL allows organizations to move existing apps onto Snowflake without code changes, according to Snowflake.

[2]

[3]

The service relies on pg_lake, a set of open source PostgreSQL extensions that allow developers and data engineers to read and write directly to Apache Iceberg tables from PostgreSQL, thereby cutting out the need to extract and move data. Iceberg is an open table format that proponents say lets users bring their preferred analytics engines to their data without moving it. It is widely used and supported across the cloud and data platform ecosystem, including by Snowflake, Google, AWS, and others.

Snowflake claims this would eliminate costly data movement between transactional and analytical systems.

[4]

The vendor already had a transactional capability within its platform. Called Unistore, it was announced in 2022, but was not generally available until late 2024. One analyst [5]commented last year that Unistore had attracted little interest.

Kleinerman said Unistore offered low-latency reads and writes, but customers were also saying they wanted a PostgreSQL-compatible service. After looking at a number of options, Snowflake bought Crunchy Data, a PostgreSQL service provider.

[6]Snowflake bets $200M that OpenAI makes databases more chatty

[7]Snowflake gets frosty with Google Gemini

[8]Snowflake buys Observe to make 'Days Since Last Outage' counters obsolete

[9]Snowflake and Databricks bank PostgreSQL acquisitions to bring transactions onto their platforms

IDC research director Devin Pratt said the move helps extend Snowflake beyond analytics into a managed OLTP offering. Online transaction processing handles live application reads and writes, while online analytical processing (OLAP) is used for reporting and analysis. Running both in one environment, he said, supports agentic AI and real-time streaming because agents need continuous access to analytical insight and live transactional data, reducing the delay between data creation and analysis.

But Snowflake was not alone in adopting the strategy. “It reflects a broader trend of vendors pairing operational databases with analytics to support real-time and agentic AI workflows,” Pratt said. Following its acquisition of Neon, which provides a serverless PostgreSQL architecture, Snowflake rival Databricks announced its service Lakebase.

In many cases, with OLTP and OLAP in the same platform, teams can reduce ETL and data duplication and apply more consistent governance and observability across transactional and analytical workloads, Pratt said.

[10]

“The value is a more unified operational and analytical stack with consistent management and security,” he said. ®

Get our [11]Tech Resources



[1] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/databases&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=2&c=2aYJ-DQQAU4P7GIN-xSDT0wAAAVg&t=ct%3Dns%26unitnum%3D2%26raptor%3Dcondor%26pos%3Dtop%26test%3D0

[2] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/databases&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aYJ-DQQAU4P7GIN-xSDT0wAAAVg&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0

[3] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/databases&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aYJ-DQQAU4P7GIN-xSDT0wAAAVg&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0

[4] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/databases&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=4&c=44aYJ-DQQAU4P7GIN-xSDT0wAAAVg&t=ct%3Dns%26unitnum%3D4%26raptor%3Dfalcon%26pos%3Dmid%26test%3D0

[5] https://www.theregister.com/2025/06/10/snowflake_and_databricks_bank_postgresql

[6] https://www.theregister.com/2026/02/02/snowflake_200m_openai/

[7] https://www.theregister.com/2026/01/06/snowflake_google_gemini_support/

[8] https://www.theregister.com/2026/01/09/snowflaketo_acquire_observe/

[9] https://www.theregister.com/2025/06/10/snowflake_and_databricks_bank_postgresql/

[10] https://pubads.g.doubleclick.net/gampad/jump?co=1&iu=/6978/reg_software/databases&sz=300x50%7C300x100%7C300x250%7C300x251%7C300x252%7C300x600%7C300x601&tile=3&c=33aYJ-DQQAU4P7GIN-xSDT0wAAAVg&t=ct%3Dns%26unitnum%3D3%26raptor%3Deagle%26pos%3Dmid%26test%3D0

[11] https://whitepapers.theregister.com/