Databricks has unveiled the general availability of Lakebase on AWS, introducing a transformative database architecture that decouples compute from storage. This managed serverless Postgres service is crafted to empower organizations to accelerate their development processes without the burden of infrastructure management.
Traditionally, databases that tightly couple compute and storage can lead to performance bottlenecks. When a single heavy query monopolizes CPU and memory resources, it can adversely impact all other operations. By adopting a model that separates these components, Lakebase allows resources to scale dynamically in response to actual workloads. When idle, the system can completely shut down, resulting in significant cost savings and minimizing unnecessary infrastructure overhead.
A standout feature of Lakebase is its instant database branching capability. This innovation enables developers to create isolated copies of production data in mere seconds, without duplicating the underlying data. As a result, teams can conduct tests on realistic datasets while safeguarding the integrity of the live environment.
Additionally, Lakebase incorporates point-in-time recovery, which offers robust protection against accidental deletions or software bugs. Organizations can revert their database to a specific millisecond within a configurable retention period, with automatic backups running seamlessly in the background.
Integration with Unity Catalog enhances Lakebase by providing consistent access control and auditing across the entire Databricks ecosystem. Sync tables ensure that operational data remains aligned with historical lakehouse context, alleviating the need for teams to manage fragile data pipelines.
Postgres 17 and scalability up to 8TB
With the launch of Lakebase, support for Postgres 17 is now included, featuring the latest enhancements and extensions, while Postgres 16 continues to be supported. The storage capacity per instance has been expanded to 8TB, accommodating larger application workloads with ease.
Databricks is also channeling its efforts into AI agents and applications through Lakebase. These applications operate directly on the Databricks platform, benefiting from shared governance, security, and the foundational data already utilized for analytics and AI. This integration eliminates the complications associated with segmented databases and disparate access controls, ensuring everything remains synchronized.
Lakebase is now available for workloads in select AWS regions, with a beta version running on Azure in specific areas. General availability on Azure is anticipated within a few months, with plans to extend to Google Cloud later this year. Compliance certifications, including SOC2 and HIPAA, are projected for early 2026.
Tip: Multi-agent systems will dominate IT environments in 2026
