What it is: For users needing to integrate MongoDB collections with analytical data warehouses, this launch introduces General Availability (GA) support for the $iceberg stage in Atlas Stream Processing to natively write Apache Iceberg v2 tables to AWS S3 buckets for use by OLAP warehouses like Snowflake or Databricks. This functionality automatically processes schema evolution, streaming inserts, updates, and deletes while integrating with AWS Glue for metadata to act as a persistent, central data catalog for robust table management and downstream platform integration.

Who it’s for: This feature is tailored for full-stack developers, principal architects, and engineering directors of all MongoDB experience levels who are tasked with capturing change streams for OLAP systems. It applies to customers of all sizes who want to eliminate the operational complexity and high infrastructure costs of running custom-coded applications or multi-stage Kafka and data warehouse connector stacks.

Why it matters: By delivering a native $iceberg stage with AWS Glue catalog support, this launch significantly improves ease of use while driving down infrastructure costs and engineering complexity. It allows organizations to establish a single, headless data source on cloud object storage that enhances interoperability across multiple analytical query engines and OLAP platforms without data duplication or pipeline sprawl.

How to get started: Get started by deploying an SP10 tier or above and use the official documentation to create your first stream processor using the $iceberg stage.