Databricks and Row64

Databricks brings data engineering, SQL warehousing and machine learning together on one lakehouse, governed by Unity Catalog and fed by streaming and batch pipelines. Row64 adds the screen your operators act from: a live command center over the same streams, without a custom app for your data team to build and maintain.

Reviewed November 2026

Keep Databricks for

  • Data and AI on one platform.

    Data engineering, SQL analytics and machine learning work from one copy of data in the lakehouse.

  • Open formats.

    Data stays in open formats such as Delta Lake, so you aren't locked in to one engine.

  • Governance with Unity Catalog.

    One set of permissions, lineage and auditing across tables, models and dashboards.

  • Streaming and batch on the same code.

    Lakeflow pipelines and Spark Structured Streaming run batch and streaming work on the same APIs.

Add Row64 for

  • Model scores on the live map.

    Scores and predictions your pipelines publish appear on the live map as they arrive, next to the asset they describe.

  • No app for the data team to maintain.

    Millions of assets on a GPU map, with drawings, photos and video beside them, configured in Row64 Studio rather than built and maintained by your data team.

  • New sources and rules, no rebuild.

    Add streams, change rules and rework views while operators are using it, with no downtime or rebuild.

  • From a score to a crew ticket.

    Rules send alerts, tickets and REST calls when a score crosses a limit, and the result comes back live.

In practice

Ice is forming on the blades. Which turbines come down first?

With Row64, a renewables operator running Databricks can watch icing spread across 2,400 wind turbines as it happens, scored by the models its data team already trains.

Before dawn, freezing fog settles over two sites. The icing scores published from Databricks climb on 90 turbines, and Row64 flags them on the map, opens tickets for the site crews and warns the grid scheduler that output will drop. The control room lead opens the worst turbines, with their power curves and blade photos, and decides which to stop.

Next week, Databricks retrains the model on the night's data.

How they work together

Keep your lakehouse. Add the screen operators act from.

01

Databricks builds it.

Pipelines, models and governed tables stay in Databricks, owned by the data team that runs them.

02

Row64 reads it.

Row64's Stream API reads the same Kafka or MQTT streams, and its Data API brings in tables from Databricks.

03

Row64 acts on it.

Alerts, tickets and REST calls go out automatically or from an operator's screen, and the result comes back live.