Broadband network gateway
Plan PPPoE or IPoE termination around subscriber identity, circuit information, address assignment and service policy. Include the handoff, VLAN/QinQ and MTU in your design.
Explore a software broadband edge that brings subscriber access, CGNAT and routing into one architectural conversation.
Plan PPPoE or IPoE termination around subscriber identity, circuit information, address assignment and service policy. Include the handoff, VLAN/QinQ and MTU in your design.
Connect private subscriber addressing to public IPv4 resources. Consider port allocation, TCP/UDP/ICMP behaviour, exhaustion and mapping traceability together.
Treat IPv4 and IPv6 as one service. Include prefix delegation, neighbour discovery, path MTU and subscriber policy in both directions.
Align forwarding intent with BGP/OSPF control through FRRouting integration. Review route policy, next-hop behaviour and traffic symmetry.
Define upload/download policy, aggregate limits, ACLs and subscriber resource limits as explicit, reviewable service intent.
Separate configuration recovery, address ownership, NAT flow continuity and subscriber session continuity. Set expectations for each failure domain.
A CGNAT decision goes beyond conserving addresses. You also need an answer when a pool fills, a subscriber consumes too many ports or an attribution query arrives.
Compare public address use, subscriber port demand, pool separation and allocation modes for your traffic mix.
Traceability needs the external address, port, protocol and timestamp, connected to allocation and subscriber history.
Separate address exhaustion, port exhaustion, state-table pressure and CPU limits in the capacity review.
Evaluate standalone CGNAT behind your existing subscriber platform. Define traffic symmetry, pool ownership and routing boundaries.
Bring access, identity and translation into a combined design. Review handoff, MTU, CPE compatibility and service policy together.
Map gateways to regional failure domains with central operational intent. Keep local resilience and site disaster recovery distinct.
Sizing starts with your workload: packet sizes, simultaneous flows, new-flow rate, subscriber churn, routes, logging and recovery headroom. Request a deployment-specific technical review.