Lakebase is a fully managed, PostgreSQL-compatible operational database built directly into the lakehouse. It brings transactional workloads and analytical data together on one governed platform, reducing the complexity and cost of running a separate operational database. Scroll to explore what sets it apart.
Create sub-second, copy-on-write clones of a database to test changes against production-like data without copying it. Branches give developers, CI pipelines, and AI agents a safe, isolated environment, and snapshots make it easy to capture and restore a known state.
Compute and storage are separated and scale independently. Instances launch in under a second, scale with demand to support high concurrency, and can scale to zero when idle, so you pay only for what you use rather than for always-on provisioned capacity.
Automatically synchronize data to and from lakehouse tables. Serve analytical data to applications with low latency through synced tables, write operational data back to the lakehouse, and power an online feature store for model serving — all without building and maintaining custom ETL.
Lakebase is built on the open-source PostgreSQL engine, so it works with the tools, drivers, and extensions your teams already use — including pgvector for embeddings. There is no proprietary dialect to learn and no lock-in to a bespoke API.
Announced at the Data + AI Summit, Lakebase supports cross-region and cross-cloud disaster recovery with one-step failover. Combined with high availability, point-in-time recovery, and encryption at rest, it provides the resilience that mission-critical applications and agents require.
Lakebase is fully managed with usage-based pricing and scale-to-zero, removing the operational burden of running database infrastructure. By unifying operational and analytical data on the lakehouse, it also eliminates the replicas and pipelines that drive up the cost of database ownership.
Traditional operational databases require dedicated infrastructure, always-on capacity, and pipelines to move data into the analytics platform. Lakebase replaces that with a fully managed service that scales to zero, charges only for what you use, and keeps operational and analytical data on a single governed copy — so teams spend less on infrastructure, data movement, and operational overhead.