BGP, OSPF, and policy routing without dedicated hardware
Tophan runs a full routing stack inside the hypervisor layer, handling inter-VLAN routing, dynamic route exchange, and policy-based path selection. This isn’t a replacement for your physical edge router — it’s the routing layer between your virtual networks, managed from the same interface as everything else.
Each Tophan node runs a FRRouting instance under Tophan’s control. Virtual routers can peer with each other over BGP, run OSPF for internal path calculation, or use simple static routes for predictable topologies. Policy routing lets you steer traffic by source, destination, application, or custom tags — useful for sending database traffic over a dedicated backend network while web traffic takes the public path.
Route tables are visible in Dragon’s Eye with real-time convergence status. When a link goes down, you see the re-convergence happen live. When you add a new subnet, you define it once and the routing fabric picks it up automatically.
For multi-site deployments connected via VPN tunnels, BGP peering between sites gives you automatic failover without manual route management.
| Feature | Description |
|---|---|
| BGP | Full eBGP/iBGP with route filtering, communities, and path prepending |
| OSPF | Area-based link-state routing for internal virtual networks |
| Static Routes | Simple next-hop definitions with failover priority |
| Policy Routing | Steer traffic by source, destination, protocol, or custom tags |
| Route Maps | Filter and modify routes on import/export between protocols |
| BFD | Bidirectional Forwarding Detection for sub-second failover |
| Equal-Cost Multi-Path | Load balance across multiple equal routes |
| Route Visualization | Live topology map with convergence status in Dragon’s Eye |