GMPLS control of ethernet

1. A method of utilizing Generalized Multi-protocol Label switching (GMPLS) to control Ethernet provider backbone transport (PBT) paths, the method comprising the steps of:

determining a path from an originating edge node to a terminating edge node through at least one intermediate node;

in either the originating edge node or the terminating edge node, comparing a GMPLS explicit route object (ERO) associated with the path to an existing set of shared forwarding multiplexes, and selecting a backbone virtual local area network identifier (VID) and a media-access-control (MAC) address in a first VID/MAC tuple which identifies what the terminating edge node considers to be an optimal shared forwarding multiplex for the originating edge node to join;

sending, from the originating edge node to the terminating edge node, a first GMPLS label for identifying the path, the first GMPLS label containing the first VTD/MAC tuple; and

using the first VID/MAC tuple for the path in a respective forwarding table of each intermediate node;

wherein when the comparing and selecting steps are performed in the terminating node, the first VIP/MAC tuple is communicated to the originating node via a GMPLS control plane; and

wherein when the comparing and selecting steps are performed in the originating node, information of the optimal shared forwarding multiplex is communicated to the originating node via the GMPLS control plane.

2. The method of claim 1 further comprising, in response to the first GMPLS label, the step of:

sending, from the terminating edge node to the originating edge node, a second GMPLS label for identifying the path, the second GMPLS label containing a second VIP/MAC tuple.

3. The method of claim 2 further comprising the step of:

installing the second VIP/MAC tuple in the respective forwarding table of each intermediate node.

4. The method of claim 2 wherein the first GMPLS label is a UPSTREAM_LABEL object and the second GMPLS label is a GENERALIZED_LABEL object in a RESV message.

5. The method of claim 1 wherein the VID/MAC tuple in the first GMPLS label comprises a 12 bit VIP and a 48 bit MAC destination address or group multicast address.

6. The method of claim 1 wherein the MAC address in the first VID/T 4 AC tuple is associated with the originating edge node.

7. An Ethernet network, utilizing Generalized Multi-protocol Label Switching (GMPLS) for establishing provider backbone transport (PBT) paths, the network comprising:

an originating edge node;

a terminating edge node; and

a plurality of intermediate nodes forming a mesh between the originating and terminating edge nodes;

wherein a path is defined between the originating edge node and the terminating edge node through at least one intermediate node, the path being identified by a GMPLS label containing a backbone virtual local area network identifier (VID) and a media-access-control (MAC) address in a VID/MAC tuple, and wherein each intermediate node of the path installs the VID/MAC tuple in a respective forwarding table to route data between the originating edge node and the terminating edge node; and

wherein the VID/MAC tuple is selected in either the originating edge node or the terminating edge node using a selection process in which a GMPLS explicit route object (ERO) associated with the path is compared to an existing set of shared forwarding multiplexes to find an optimum shared forwarding multiplex for the originating edge node to join, and the VID/HAC tuple identifying what the terminating edge node considers to be the optimal shared forwarding multiplex;

wherein when the comparing and selecting steps are performed in the terminating node, the VID/MAC tuple is communicated to the originating node via a GMPLS control plane; and

wherein when the comparing and selecting steps are performed in the originating node, information of the optimal shared forwarding multiplex is communicated to the originating node via the GNPLS control plane.

8. The network of claim 7 further comprising a path computation server for determining the path between the originating edge node and the terminating edge node through the at least one intermediate node in a GMPLS network topology.

9. The network of claim 8 wherein the GMPLS topology is determined by a Link Management Protocol (LMP) using 802.1ab extensions.

10. The network of claim 7 wherein the VID/MAC tuple is associated with a GMPIJS port index of a selected one of the originating edge node and the terminating edge node.

11. A method performed by an Ethernet bridging node in a mesh network, the method comprising steps of:

receiving a first Generalized Multi-protocol Label Switching (GMPLS) label from a first edge node, the first GMPLS label identifying a provider backbone transport (PBT) path through the mesh network between the first edge node and a second edge node, the first GMPLS label containing a backbone virtual local area network identifier (VID) and a media-access-control (MAC) address associated with the first edge node in a VID/MAC tuple;

using the first GMPLS label VID/MAC tuple in a forwarding table of the Ethernet bridging node, the first GMPLS label VID/MAC tuple identifying an egress port of the Ethernet bridging node associated with the PBT path, wherein packets received at an ingress port of the Ethernet bridging node are forwarded to the egress port of the Ethernet bridging node based on VID/MAC tuples in the packets; and

forwarding the first GMPLS label to a next node along the PBT path toward the second edge node of the PBT path;

wherein the first GMPLS label VID/MAC tuple is selected in either the first edge node or the second edge node using a selection process in which a GMPLS explicit route object (ERO) associated with the PBT path is compared to an existing set of shared forwarding multiplexes to find an optimum shared forwarding multiplex for the first edge node to join, the first GMPLS label VID/MAC tuple identifying what the second edge node considers to be the optimal shared forwarding multiplex;

wherein when the comparing and selecting steps are performed in the second edge node, the first GMPLS label VID/MAC tuple is communicated to the first edge node via a GMPLS control plane; and

wherein when the comparing and selecting steps are performed in the first edge node, information of the optimal shared forwarding multiplex is communicated to the first edge node via the GMPLS control plane.

12. The method of claim 11 wherein a second GMPLS label is received from the second edge node, the first and second GMPLS labels identifying respective unique first and second VID/MAC tuples for forwarding packets to the associated edge node.

13. The method of claim 12 wherein the first and second VID/MAC tuples are installed in the forwarding table of the Ethernet bridging node such that packets received at ingress ports of the Ethernet bridging node are forwarded along the PBT path by the egress ports based upon the VID/MAC tuple in each of the packets.

14. The method of claim 11 wherein the first GMPLS label is received at the Ethernet bridging node in an UPSTREAM_LABEL object or a GENERALIZED_LABEL object.