Architecture

Fyrefuse architecture can be deployed and run in a Self-managed setup or leverage Kubernetes orchestration capabilities to support larger data processing capacity.

Components

Fyrefuse is built on a modular architecture that combines three core technologies: Apache Spark for distributed data processing, Trino for federated SQL querying, and a Python-based backend that coordinates the pipeline logic. Together, these components enable the development of powerful and flexible data workflows. For pipelines to function correctly, all three dependencies must be installed on the host system.

Next steps

• Learn more about Fyrefuse architecture discovering it's ecosystem.
• Learn how to easily setup our software with our default installation, before trying a custom installation.