Plan a pod
Menu
Platform

Lakebase and LTAP: one copy of data for apps and analytics

Lakebase brings managed Postgres into the lakehouse, and LTAP aims to serve transactional and analytical work from a single copy of data.

By NeuralShore EngineeringPublished 1 min read
Cover illustration: Lakebase and LTAP: one copy of data for apps and analytics
Short answer

Lakebase is Databricks’ managed Postgres inside the governed lakehouse, and LTAP aims to serve transactional and analytical work from one copy of data. Together they reduce the copying between app databases and the warehouse. Check which capabilities are generally available in your region before you design around them.

Key takeaways

  • Lakebase runs Postgres under Unity Catalog governance.
  • Fewer copies means fewer pipelines and less drift.
  • Good fits: apps, agents and features that need fresh data.

Most companies still run two worlds. Applications write to an operational database. Then pipelines copy that data into a warehouse or lakehouse before anyone can analyse it.

Databricks is trying to close that gap.

Lakebase: Postgres inside the lakehouse

Lakebase is a managed Postgres service on Databricks. This year it gained cross-cloud disaster recovery, branching and more autonomous operations. Lakebase Search adds hybrid vector and full-text search in beta.

For teams building internal apps or AI agents, this matters. The app's own database sits next to governed analytical data, under the same security model.

LTAP: one copy, two workloads

Databricks also announced LTAP, short for Lake Transactional Analytical Processing. The goal is to serve both operational and analytical work from one copy of data. It was announced as coming soon, so it is not ready for production plans yet.

What we recommend today

  • Use Lakebase for new internal apps, agent memory and operational features that sit close to analytics.
  • Keep existing, well-run pipelines in place. Do not rebuild them around a feature that is not generally available.
  • Design new data models so they could move to a single-copy pattern later.

The bottom line

Fewer copies mean fewer pipelines to break and fewer places for data to drift. Lakebase is a practical step now. LTAP is worth watching for the next planning cycle.

About the author

NeuralShore Engineering, Databricks delivery team

Databricks architects, data engineers and AI engineers at NeuralShore, DLytica’s Databricks delivery company. Leadership comes from IBM data and AI architecture.

FAQ

Frequently asked questions

What is Databricks Lakebase?

Lakebase is a managed Postgres database that runs inside the Databricks platform under Unity Catalog governance.

When should we use Lakebase?

For apps, agents and features that need fresh operational data next to analytics, without copying data between systems.

Keep reading

Related articles