What it is: Atlas Stream Processing now includes native support for Google Cloud Pub/Sub as a sink. This feature allows developers to stream data directly from MongoDB through Atlas Stream Processing and publish it to Pub/Sub topics.

Who it’s for: This is for developers and data engineers who utilize the Google Cloud ecosystem (e.g. BigQuery, Dataflow, Cloud Functions) and want to integrate their real-time operational data from MongoDB without complex infrastructure.

Why it matters: The direct integration removes the need for separate streaming infrastructure, significantly simplifying real-time data pipeline architectures. It enables key use cases like powering agentic AI with real-time context, enabling real-time analytics in BigQuery, and building responsive event-driven architectures based on data changes in MongoDB.

How to get started:

1. Configure GCP Service Account: Grant the Atlas Service Account provided in the Project Integrations section of Atlas the necessary roles/pubsub.viewer and roles/pubsub.publisher permissions in your Google Cloud Project.

2. Define Connection: Add Google Cloud Pub/Sub as a connection in the Atlas Connection Registry.
3. Stream Data: Use the $emit stage in your Atlas Stream Processor to target the Pub/Sub connection, specifying the topic, project ID, and optionally, custom attributes and ordering keys.