Distributed Architecture
AI Hydra components run independently and can be deployed across different machines.
Example Deployment
This configuration demonstrates a fully distributed setup where simulation, routing, and visualization are physically separated across hosts. This example also uses custom ports. The Hydra Client is running locally, the Hydra Router is running on bingo, and the Hydra Manager is running on islands.
1. Client
# Client (local machine)
$ . hydra-venv/bin/activate
hydra-venv> ai-hydra-client --router-port 6757 --router-hb-port 6758 --server-pub-port 6760 --router-address bingo --server-address islands
2. Router
# Router (bingo)
$ . hydra-venv/bin/activate
hydra-venv> ai-hydra-router --port 6757 --hb-port 6758
3. Server
# Manager / Simulation (islands)
$ . hydra-venv/bin/activate
hydra-venv> ai-hydra-mgr --port 5759 --router-port 6757 --router-hb-port 6758 --router-address bingo