Run governed Kafka across on-premises, hybrid, and cloud environments without giving up control of your data.
Databases, message queues, cloud storage, and legacy enterprise systems, including the mainframes and older protocols that cloud-only platforms leave out.
Deploy through the same self-service model you already use for streaming applications: UI, REST API, or Terraform. No ticket, no waiting on a platform team.
Connector credentials are stored in HashiCorp Vault, while TLS, isolated workers, and role-based access control keep deployments secure.
.png)
Source connectors pull data from external systems into Kafka, while sink connectors push Kafka data to external destinations.
A connector is the structure that enables data flow between an external system and Kafka. That external system could be a database, key-value store, search index, file system, or similar component — connectors are ready-to-use, so you don't need to write custom integration code for most common systems.
Both. Axual supports its own pre-built connectors and lets you bring your own custom connectors into the same managed framework.
All connectors are included as part of the platform — there's no extra fee to add integrations across the data sources and systems Axual supports.
Every connector's credentials are stored in HashiCorp Vault, never in plain configuration. Connectors run in isolated worker clusters with TLS encryption, and role-based access control ensures only authorized users can create connectors or access specific data sources.
Through the same self-service model used for streaming applications: deploy via the UI, REST API, or Terraform. Connector ownership is registered like any other application, so your team manages the full connector lifecycle — including inspecting logs to debug failures — without filing a ticket.
Yes. Connectors run on Kafka Connect, which scales by adding workers or tasks to handle increased load, and supports standard transformations (filtering, masking, routing) on messages as they move through.
No — connectors run in isolated worker clusters, separate from your Kafka brokers, so onboarding a new integration doesn't compete with your core streaming workloads for resources.