Redpanda 25.2 release simplifies Iceberg adoption for enterprise users building real-time AI/ML pipelines

Redpanda, the real-time data platform that powers agentic and operational applications, today announced Redpanda 25.2, which simplifies Iceberg adoption for enterprise users. The latest update also includes support for Databricks Unity Catalog, allowing enterprise customers to turn streaming event data into queryable Iceberg tables in Databricks with unprecedented ease and control.

By integrating Redpanda Iceberg Topics and Databricks Unity Catalog, streaming data becomes immediately queryable in Databricks using Spark or DB SQL as Managed Iceberg tables. This eliminates the need for batch ETL, enabling AI/ML models, dashboards, and data pipelines to consume fresh, real-time data in a unified, open lakehouse architecture.

“Apache Iceberg represents the future of not just streaming, but enterprise data infrastructure as a whole,” said Tyler Akidau, CTO of Redpanda. “At Redpanda, we’ve been committed to advancing Iceberg as a foundational element of modern data infrastructure, and the 25.2 release continues that momentum. Our new partnership with Databricks is especially timely as Iceberg V3 gains momentum to unify lakehouse data formats across Delta Lake, Apache Parquet, and Apache Spark.”

In addition to support for Databricks Unity Catalog, the 25.2 release brings multiple new integrations and features aimed at improving flexibility, security, and developer experience:

• AWS Glue Support: Redpanda has expanded its Iceberg Catalog support with AWS Glue as an Iceberg REST catalog, making it easy for AWS-centric teams to use Glue as a central metadata repository for all data artifacts, including Redpanda Iceberg Topics. This enables native integration with AWS analytics and AI tools without introducing additional lakehouse infrastructure or requiring synchronization between multiple catalogs.
• JSON Schema Support for Iceberg: Redpanda 25.2 adds support for JSON schemas within Iceberg Topics, joining Avro and Protobuf.
• Schema Registry Authorization: New fine-grained access control (ACLs) for Redpanda’s Schema Registry improves security and governance. Users can now assign schema and subject-level permissions to roles and individuals, aligning with enterprise-grade compliance and governance models.
• Support for kafka-javascript: Redpanda has certified the modern Node.js client kafka-javascript, ensuring developers building real-time services with JavaScript can do so with confidence and native compatibility.

To learn more about Redpanda 25.2, read our launch blog post.

About Redpanda

Redpanda is a real-time data platform that powers agentic and operational applications by integrating real-time event processing, historical analytics, and AI connectivity into a unified system. Unlike traditional architectures that require sending data to external AI models, Redpanda and its innovative BYOC capabilities brings the model to the data. Global leaders including Activision Blizzard, Cisco, Moody's, Texas Instruments, Vodafone and 2 of the top 5 banks in the U.S. rely on Redpanda to process hundreds of terabytes of data a day. Backed by premier venture investors Lightspeed, GV and Haystack VC, Redpanda is a diverse, people-first organization with teams distributed around the globe. To learn more, visit our website at redpanda.com and follow us on X (formerly Twitter) at @redpandadata.

View source version on businesswire.com: https://www.businesswire.com/news/home/20250805924617/en/