Granted Patent B2

US 8,934,501 · App. 13/493,985 · Granted Jan 13, 2015

Abstract

Claims (54)

1. A method for bidirectional translation of edge network virtualization encapsulation to core network virtualization encapsulation, the method comprising:

receiving a multicast customer data frame at a first switch, the customer data frame having a first multicast edge network virtualization encapsulation header;

removing the first multicast edge network virtualization encapsulation header from the customer data frame;

adding a multicast core network virtualization encapsulation header to the customer data frame;

forwarding the message to a second switch according to a mapping obtained from the first switch, wherein the mapping includes a virtual system MAC address and a physical system MAC address for a network identifier of a network associated with a sender of the customer data frame;

receiving the customer data frame at the second switch;

removing the multicast core network virtualization encapsulation header from the customer data frame;

adding a second multicast edge network virtualization encapsulation header to the customer data frame; and

forwarding the customer data frame according to the second multicast edge network virtualization encapsulation header.

2. The method of claim 1 , further comprising:

learning, at the first switch, a mapping between a header of the customer data frame and an outer edge network virtualization encapsulation header associated with a sender of the customer data frame.

3. The method of claim 1 , further comprising:

learning, at the second switch, a mapping between a source of the customer data frame and first switch data from the core network virtualization encapsulation header for a service identifier associated with the first switch.

4. The method of claim 1 , wherein the edge network virtualization encapsulation header includes a VXLAN header, and the core network virtualization encapsulation header includes a MAC-in-MAC encapsulation header.

5. The method of claim 1 , wherein the edge network virtualization encapsulation header includes a NVGRE header, and the core network virtualization encapsulation header includes a MAC-in-MAC encapsulation header.

6. The method of claim 1 , wherein the first switch includes a mapping from a service identifier present in the edge network virtualization encapsulation header of the customer data frame to a service identifier in the core network virtualization encapsulation header.

7. The method of claim 6 , wherein the mapping is static and is manually configured.

8. The method of claim 6 , wherein the mapping is dynamic and is automatically learned.

9. A system comprising:

a first processor coupled to a first nontransitory computer readable medium having stored thereon software instructions that, when executed by the first processor, cause the first processor to perform a first series of operations, the first series of operations including:

receiving a unicast customer data frame at a first switch, the customer data frame having a first unicast edge network virtualization encapsulation header;

removing the first unicast edge network virtualization encapsulation header;

adding a unicast core network virtualization encapsulation header to the customer data frame;

forwarding the message to a second switch according to a mapping obtained from the first switch, wherein the mapping includes a virtual system MAC address and a physical system MAC address for a network identifier of a network associated with a sender of the customer data frame; and

a second processor coupled to a second nontransitory computer readable medium having stored thereon software instructions that, when executed by the second processor, cause the second processor to perform a second series of operations, the second series of operations including:

receiving the customer data frame at the second switch;

removing the unicast core network virtualization encapsulation header from the customer data frame;

adding a second unicast edge network virtualization encapsulation header to the customer data frame; and

forwarding the customer data frame according to the second edge network virtualization encapsulation header.

10. The system of claim 9 , wherein the first series of operations further comprises:

learning, at the first switch, a mapping between a header of the customer data frame and an outer edge network virtualization encapsulation header associated with a sender of the customer data frame.

11. The system of claim 9 , wherein the second series of operations further comprises:

learning, at the second switch, a mapping between a source of the customer data frame and first switch data from the core network virtualization encapsulation header for a service identifier associated with the first switch.

12. The system of claim 9 , wherein the edge network virtualization encapsulation header includes a VXLAN header, and the core network virtualization encapsulation header includes a MAC-in-MAC encapsulation header.

13. The system of claim 9 , wherein the edge network virtualization encapsulation header includes a VLAN header, and the core network virtualization encapsulation header includes a MAC-in-MAC encapsulation header.

14. The system of claim 9 , wherein the first switch includes a mapping from a service identifier present in the edge network virtualization encapsulation header of the customer data frame to a service identifier in the core network virtualization encapsulation header.

15. The system of claim 14 , wherein the mapping is static and is manually configured.

16. The system of claim 14 , wherein the mapping is dynamic and is automatically learned.

17. A nontransitory computer readable medium having stored thereon software instructions that, when executed by a processor, cause the processor to perform a series of operations comprising:

receiving a customer data frame at a first switch, the customer data frame having a first edge network virtualization encapsulation header;

removing the first edge network virtualization encapsulation header;

determining whether the customer data frame was being sent as a unicast message or a multicast message based on the first edge network virtualization encapsulation header;

adding a unicast core network virtualization encapsulation header to the customer data frame when the customer data frame was being sent as a unicast message;

adding a multicast core network virtualization encapsulation header to the customer data frame when the customer data frame was being sent as a multicast message;

forwarding the message to a second switch according to a mapping obtained from the first switch, wherein the mapping includes a virtual system MAC address and a physical system MAC address for a network identifier of a network associated with a sender of the customer data frame;

receiving the customer data frame at the second switch;

removing, at the second switch, the core network virtualization encapsulation header;

adding an edge network virtualization encapsulation header to the customer data frame; and

forwarding the customer data frame according to the edge network virtualization encapsulation header.

18. The nontransitory computer readable medium of claim 17 , wherein the operations further comprise:

learning, at the first switch, a mapping between a header of the customer data frame and an outer edge network virtualization encapsulation header associated with a sender of the customer data frame.

19. The nontransitory computer readable medium of claim 17 , wherein the operations further comprise:

learning, at the second switch, a mapping between a source of the customer data frame and first switch data from the core network virtualization encapsulation header for a service identifier associated with the first switch.

20. The nontransitory computer readable medium of claim 17 , wherein the edge network virtualization encapsulation header includes a VXLAN header, and the core network virtualization encapsulation header includes a MAC-in-MAC encapsulation header.

Assignments (13)