Snowflake Postgres is Now Available in Public Preview

Today, we are thrilled to announce the public preview availability of Snowflake Postgres, our fully managed Postgres offering. This release brings the world’s most popular database, Postgres, onto Snowflake. With Snowflake Postgres, you get a new set of capabilities that support a wide array of operational workloads within the Snowflake AI data platform, making it easier than ever to leverage your data for advanced AI, ML and analytics applications with simpler management, security and governance.

For years, our customers have told us they love Postgres, whether they’re app developers, analysts or data engineers. Love for Postgres is pervasive, and for good reason. Postgres is reliable, always putting an emphasis on data safety and consistency. Postgres is flexible; it’s a Swiss Army knife of a database, with a solution to almost any problem. And Postgres is extensible. With its extension framework, Postgres’ capabilities become a question of imagination. As an example, we open sourced pg_lake last month, which lets you easily work with data lakes from Postgres natively.

What's included in Snowflake Postgres?

Snowflake Postgres has everything you need to get started today, and scale your apps tomorrow, while resting easy at night knowing your data will be safe, secure, and ready to support whatever your business needs. Today’s preview release includes a set of enterprise features supporting a wide range of use cases, including:

And more capabilities are coming soon.

Postgres in the Snowflake AI Data Cloud

Postgres is a 30-year-old, battle-tested technology with leading reliability and performance, but it comes with a more classical architecture that works differently from the rest of Snowflake. We needed to figure out how to make Postgres work on the Snowflake platform.

Snowflake Postgres provides a running Postgres instance with attached disks, giving you best- in-class database performance. This differs from the Snowflake virtual warehouse model. Each Snowflake Postgres instance runs in its own isolated private network and supports private connectivity via firewall rule or PrivateLink and, coming soon, VPC peering. Snowflake Postgres also comes with built-in connection pooling via pg_bouncer. And we’re working to provide increasingly seamless data movement between Postgres and the rest of Snowflake, making it easier than ever to use Snowflake and Postgres together.

Snowflake Postgres builds upon the proven technology that we acquired from Crunchy Data, and is operated by a team with years of experience running a Postgres service in the cloud. This provides a production-ready management layer on top of the open source Postgres you know and love. Snowflake Postgres is compatible with all the Postgres tooling you already use because it's not a fork of Postgres, meaning you can easily ‘lift and shift’ your existing Postgres apps onto Snowflake with no code changes. 

Get started today 

To get started with Snowflake Postgres, just click Postgres in your Snowsight navigation bar, click Create, then configure and provision your instance. That’s it. You’ll get credentials and can connect to your instance from any Postgres client.