Understand this story
On November 13, 2026, Databricks launched Lakehouse//RT, enabling real-time analytics to run directly on the lakehouse, with costs that do not grow proportionally with query volume. According to the Databricks Blog, this engine can deliver sub-100-millisecond latency at 12,000 QPS and supports small-step scaling by single node. Customers who chose it to replace a separate real-time serving layer report infrastructure cost reductions of 6x to 10x. Customer Aurasell migrated from multiple clusters at 200 QPS to a single small Lakehouse//RT warehouse, reliably supporting over 250 QPS, with query speed improved 4-5x and costs reduced 10x. There are currently no subsequent developments.
AI-generated from public reports
Latest development · Report
Databricks Lakehouse//RT: Scaling Real-Time Analytics Without Inflating the BillDevelopments
1 confirmed developments
Databricks launches Lakehouse//RT real-time analytics warehouse
Source:Databricks Blog
Reports and discussion timeline
Browse reports and discussion by time. Their evidence levels remain distinct.
Nov 13
Databricks BlogFirst-party
Databricks Lakehouse//RT: Scaling Real-Time Analytics Without Inflating the Bill
Databricks introduces Lakehouse//RT, which lets real-time analytics run directly on the lakehouse, with costs that do not grow proportionally with query volume; customers who chose it to replace a separate real-time serving layer report infrastructure cost reductions of 6x to 10x. The engine delivers sub-100-millisecond latency at 12,000 QPS and supports small, single-node incremental scaling. Aurasell migrated from multiple clusters at 200 QPS to a single small Lakehouse//RT warehouse, reliably sustaining over 250 QPS, with queries 4-5x faster and costs reduced by 10x.