gpu-fleet
Systems ReplayControl plane for bare-metal GPU clouds: node lifecycle over Redfish, topology-aware capacity reservations, tenant attestation and automated remediation. Go control plane, Rust node agent, Postgres. A chaos service keeps a 50-node simulated fleet breaking and healing itself, and the project page replays a simulated run in the browser.
Fault to quarantine in 88 s, replacement 2 s later
- Go
- Rust
- PostgreSQL
- Backend