Google Cloud announced new integrations between Gemini and its borderless Lakehouse on October 15, 2026. The update enables AI agents and business users to access and query data across multiple clouds using natural language without data movement.
Gemini Lakehouse Integration
According to Google Cloud, data preconfigured or stored in the borderless Lakehouse is now directly integrated into Gemini. Currently in preview, users can set up access to new data sources, such as Databricks Unity Catalog or Snowflake Horizon, through Lakehouse catalog federation directly from Gemini.
Google Cloud stated that business users can use natural language to query records living across clouds, while AI agents ground their reasoning and generate code from the same data without moving it.
General Availability and Performance Features
Google Cloud announced that several borderless Lakehouse features are now generally available. These include support for Cross-Cloud Interconnect and cross-cloud caching for AWS and Azure, as well as BigQuery support for Apache Iceberg V3 specifications.
The update also brings read-write interoperability on Apache Iceberg tables for BigQuery and Managed Service for Apache Spark. Additionally, Google Cloud added zero-copy data integration with Confluent Tableflow, Salesforce, SAP, and Workday.
For organizations using legacy Dataproc Metastore, Google Cloud now offers an automated, zero-downtime migration path to the serverless Lakehouse runtime catalog.
Key facts and where they come from
- The Gemini integration with the borderless Lakehouse is currently in preview.
Currently in preview, this means you can set up access to new data sources, such as Databricks Unity Catalog or Snowflake Horizon, through Lakehouse catalog federation directly from Gemini.
- BigQuery now supports Apache Iceberg V3 specifications.
BigQuery support for Apache Iceberg V3 specifications
- Google Cloud provides an automated migration path from legacy Dataproc Metastore to the serverless Lakehouse runtime catalog.
For organizations currently using legacy Dataproc Metastore, we now offer an automated, zero-downtime migration path to the serverless Lakehouse runtime catalog.
