---
title: "Introducing Lakehouse//RT: Real-Time Performance on a Unified Lakehouse"
description: "Databricks introduces Lakehouse//RT, a real-time data warehouse designed for operational analytics, BI, app serving, and observability workloads. It is powered by Reyden and is intended to deliver millisecond performance directly on lakehouse data without copying it into a separate serving layer. Preview participants saw up to 16x better performance, with response times as low as 10ms on smaller datasets, sub-100ms on larger ones, and sub-100ms latency at 12,000 queries per second on standard analytical benchmarks. Tests covering concurrency, dataset scale, and complex TPCDS queries report low latency where alternatives slowed or failed. Lakehouse//RT is in Beta for select read-only workloads, with incremental autoscaling and automatic baseline compute sizing."
---

# Introducing Lakehouse//RT: Real-Time Performance on a Unified Lakehouse

[Databricks](https://yomu.fyi/company/databricks) · Nong Li, Shoumik Palkar, Shant Hovsepian, Mostafa Mokhtar, Reynold Xin · Jun 16, 2026

**Type:** Announcement

## Summary

Databricks introduces Lakehouse//RT, a real-time data warehouse designed for operational analytics, BI, app serving, and observability workloads. It is powered by Reyden and is intended to deliver millisecond performance directly on lakehouse data without copying it into a separate serving layer. Preview participants saw up to 16x better performance, with response times as low as 10ms on smaller datasets, sub-100ms on larger ones, and sub-100ms latency at 12,000 queries per second on standard analytical benchmarks. Tests covering concurrency, dataset scale, and complex TPCDS queries report low latency where alternatives slowed or failed. Lakehouse//RT is in Beta for select read-only workloads, with incremental autoscaling and automatic baseline compute sizing.

## Context

Separate real-time serving layers require copying data into proprietary storage, creating additional ingestion pipelines, operational overhead, governance duplication, and engineering work. The source says these layers can also struggle with complex queries, larger datasets, and high concurrency.

## Approach / What changed

Lakehouse//RT brings real-time performance directly to the lakehouse through the Reyden engine, using the existing open formats, governance model, and central data architecture without data movement. It automatically sizes baseline compute and uses incremental autoscaling by adding or removing individual nodes as demand changes.

## Takeaways

- Lakehouse//RT is designed to support operational analytics, BI, app serving, and observability on a single copy of lakehouse data.
- The reported tests measured performance under increasing concurrency, dataset sizes up to a terabyte using TPC-H, and complex TPCDS queries involving joins, subqueries, and window functions.
- The Beta supports select read-only workloads and includes automatic baseline compute sizing plus incremental autoscaling rather than doubling or tripling whole warehouse copies.

**Tags:** [Lakehouse](https://yomu.fyi/topic/lakehouse), [Performance](https://yomu.fyi/topic/performance), [Scalability](https://yomu.fyi/topic/scalability)

- Source: [Databricks](https://www.databricks.com/blog/introducing-lakehousert-real-time-performance-unified-lakehouse)
- Source URL: https://www.databricks.com/blog/introducing-lakehousert-real-time-performance-unified-lakehouse
- Ingested by Yomu: 2026-08-30T17:02:36.443Z

[Read original post](https://www.databricks.com/blog/introducing-lakehousert-real-time-performance-unified-lakehouse)