A data model for Apache Kafka has existed in the CMDB for many years. However, the data model was originally designed primarily for on-prem installations where things like configuration files and install directories are important. Since that time there have been a number of commercial offerings for Kafka which have gained in popularity and widely adopted by customers.
One such organization is Confluent, which was purchased by IBM in early 2026. Confluent offers a variety of enterprise support for Kafka, one of which is their software-as-a-service option. To facilitate discovery of this environment in a consistent way as other cloud-based resources (like AWS, Azure, etc) ServiceNow has updated the data model with new classes as well as a number of updates to existing classes and relationships. The below diagram represents the updates that will be released in the September 2026 version of the CMDB class models app.
Updated Kafka data model
In coming releases there are plans to continue enhancing the model to also capture users and/or service accounts and their corresponding access to specific topics so that better impact may be calculated when issues arise with a topic or set of topics availability. Likewise, a Service Graph Connector for Confluent (cloud) is planned to be released in the September 2026 timeframe in conjunction with this updated data model.
The classes in this data model which will be targeted by this new SGC are the Confluent Organization, Confluent Environment, Kafka Cluster, and Kafka Topic classes. As we look past the v1.0 version, plans exist to add imports for other entities, but please let us know what would be important to your organization.
