Provider backbone bridging—provider backbone transport internetworking
An Ethernet virtual switched sub-network (VSS) is implemented as a virtual hub and spoke architecture overlaid on hub and spoke connectivity built of a combination of Provider Backbone Transport (spokes) and a provider backbone bridged sub-network (hub). Multiple VSS instances are multiplexed over top of the PBT/PBB infrastructure. A loop free resilient Ethernet carrier network is provided by interconnecting Provider Edge nodes through access sub-networks to Provider Tandems to form Provider Backbone Transports spokes with a distributed switch architecture of the Provider Backbone Bridged hub sub-network. Provider Backbone transport protection groups may be formed from the Provider Edge to diversely homed Provider Tandems by defining working and protection trunks through the access sub-network. The Provider Backbone Transport trunks are Media Access Control (MAC) addressable by the associated Provider Edge address or by a unique address associated with the protection group in the Provider Backbone Bridged network domain.
1. A method of forwarding packets at a hub of a hub and spoke packet network, the method comprising:
configuring at least two connections between the hub and an edge node;
configuring the at least two connections as a protection group such that one of the connections of the protection group is configured as a primary connection to carry packets between the hub and the edge node when the primary connection is operational and at least one other of the connections of the protection group is configured as a standby connection to carry packets between the hub and the edge node when the primary connection is not operational;
associating a protection group address with the protection group;
learning forwarding information at the hub based on source addresses of packets received at the hub, packets received from the edge node on the primary connection of the protection group having the protection group address as a source address; and
at the hub, forwarding packets having the protection group address as a destination address onto an operational connection of the protection group based on the learned forwarding information, whereby packets are forwarded at the hub onto the standby connection of the protection group based on the protection group address following failure of the primary connection without requiring relearning of forwarding information at the hub.
2. The method of claim 1 , wherein the hub comprises at least two provider tandem nodes, one of the provider tandem nodes being a primary provider tandem node connected to the edge node via the primary connection of the protection group and another of the provider tandem nodes being a standby provider tandem node connected to the edge node via the standby connection of the protection group.
3. The method of claim 2 , wherein the edge node determines which connection of the protection group is the primary connection.
4. The method of claim 2 , wherein the primary provider tandem node is configured to forward packets having the protection group address as a destination address between the hub and the edge node when the primary connection and the primary provider tandem node are operational and the standby provider tandem node is configured to forward packets between the hub and the edge node when at least one of the primary connection and the primary provider tandem node is not operational.
5. The method of claim 4 , wherein the hub comprises at least one other provider tandem node configured to forward packets addressed to the protection group via the primary provider tandem node and the primary connection based on the protection group address when the primary provider tandem node and the primary connection are operational.
6. The method of claim 5 , wherein the at least one other provider tandem node is configured to forward packets addressed to the protection group via the standby tandem node and the standby connection based on the protection group address when at least one of the primary provider tandem node and the primary connection is not operational without requiring relearning of forwarding information at the at least one other hub.
7. The method of claim 6 , wherein, when at least one of the primary connection and the primary provider tandem node fails, the standby provider tandem node is configured to send a broadcast packet on the hub with the protection group address as a source address to inform other nodes of the hub to forward traffic for the protection group address to the standby provider tandem node.
8. The method of claim 5 , wherein the hub is a provider backbone bridging (PBB) network.
9. The method of claim 8 , wherein packets addressed to the protection group are forwarded over the PBB network based on a combination of a backbone virtual network identifier and a backbone destination address, the backbone destination address being the protection group address.
10. The method of claim 9 , wherein packets received at the hub from the edge node on either the primary connection or the standby connection are forwarded over the PBB network with the protection group address as a backbone source address.
11. The method of claim 1 , wherein the at least two connections are provider backbone transport (PBT) trunks.
12. The method of claim 11 , wherein the primary connection comprises a primary PBT trunk and packets are forwarded between the hub and the edge node over the primary PBT trunk with a backbone virtual network identifier associated with the primary trunk.
13. The method of claim 11 , wherein the standby connection comprises a standby PBT trunk and packets are forwarded between the hub and the edge node over the standby PBT trunk with a backbone virtual network identifier associated with the standby trunk.
14. The method of claim 1 , wherein packets forwarded between the hub and the edge node comprise service identifiers and packets are forwarded at the hub based on mappings selected based on the service identifiers.
15. The method of claim 14 , wherein the mappings associate customer addresses with backbone addresses.
16. The method of claim 15 , wherein the mappings are associated with virtual switch instances.
17. The method of claim 15 , wherein some of the mappings associate customer addresses with protection group addresses.
18. The method of claim 15 , wherein some of the protection group addresses are respective edge node addresses.
19. A method of providing a resilient attachment to a provider backbone bridged (PBB) network, the method comprising:
configuring at least two connections between the PBB network and an external node external to the PBB network;
configuring the at least two connections as a protection group such that one of the connections of the protection group is configured as a primary connection to carry packets between the PBB network and the external node when the primary connection is operational and at least one other of the connections of the protection group is configured as a standby connection to carry packets between the PBB network and the external node when the primary connection is not operational;
associating a protection group address with the protection group;
learning forwarding information at the PBB network based on source addresses of packets received at the PBB network, packets received from the external node on a connection of the protection group having the protection group address as a source address; and
forwarding packets having the protection group address as a destination address onto an operational connection of the protection group based on the learned forwarding information, whereby packets are forwarded at the PBB network onto the standby connection of the protection group based on the protection group address following failure of the primary connection without requiring relearning of forwarding information at the PBB network.
20. The method of claim 19 , wherein the external node determines which connection of the protection group is the primary connection.
21. The method of claim 19 , wherein:
the PBB network comprises at least two PBB nodes, one of the PBB nodes being a primary node connected to the external node via the primary connection of the protection group and another of the PBB nodes being a standby node connected to the edge node via the standby connection of the protection group; and
the standby node is configured, when at least one of the primary connection and the primary node fails, to send a broadcast packet on the PBB network with the protection group address as a source address to inform other nodes of the PBB network to forward traffic for the protection group address to the standby node.