Expanding network functionalities for openflow based split-architecture networks

• US 8,971,338 B2
• Filed: 01/09/2012
• Issued: 03/03/2015
• Est. Priority Date: 01/09/2012
• Status: Active Grant

• Current Assignee: Telefonaktiebolaget LM Ericsson
• Original Assignee: Telefonaktiebolaget LM Ericsson
• Inventors: Mishra, Ramesh, Beliveau, Ludovic
• Primary Examiner(s): Cho, Un C
• Assistant Examiner(s): COSTIN, JEREMY M
• Application Number:
  US13/346,073
• Publication Number:
  US 20130176850A1
• Time in Patent Office: 1,149 Days
• Field of Search: 370/401, 370/402, 370/465, 370/466, 370/467, 370/469, 709/201
• US Class Current: 370/401
• CPC Class Codes: H04L 12/46 Interconnection of networks
• CPC Class Codes: H04L 12/66 Arrangements for connecting between networks having differing types of switching systems, e.g. gateways
• CPC Class Codes: H04L 45/38 Flow based routing
• CPC Class Codes: H04L 49/00 Packet switching elements

Abstract

A method enables a switch in a split-architecture network to provide high speed packet processing and enhanced network functionalities that are not supported by the OpenFlow. The switch receives a packet from a network through an input port of the switch, and matches header fields in the packet against table entries in flow tables to identify an action to be taken. The flow tables are part of an OpenFlow pipeline. The identified action is to direct the packet to a designated processing unit in the switch. The OpenFlow pipeline forwards the packet to the designated processing unit via a communication channel in the switch. The designated processing unit processes the packet with the enhanced network functionalities, and injects the packet back to one of the flow tables before transmission of the packet to the network through an egress port of the switch.

Claims

1. A method implemented by a network element functioning as a switch in a split-architecture network that includes a controller coupled to a plurality of switches, the method enabling the switch to provide high speed packet processing and enhanced network functionalities, the method comprising the steps of:

receiving a packet by the switch from a network through an input port of the switch;

matching one or more header fields in the packet against a set of table entries in a set of flow tables to identify an action to be taken, the set of flow tables being part of an OpenFlow pipeline;

identifying the action in a matching table entry as directing the packet to a designated one of a set of processing units in the switch, wherein the set of processing units in the switch are external to the OpenFlow pipeline;

forwarding the packet to the designated processing unit from the OpenFlow pipeline via a communication channel in the switch;

processing the packet by the designated processing unit with the enhanced network functionalities, wherein the enhanced network functionalities include functionalities that are not supported by the OpenFlow pipeline of the switch; and

injecting the packet from the designated processing unit back to one of the flow tables in the OpenFlow pipeline before transmission of the packet to the network through an egress port of the switch.

2. The method of claim 1, wherein the step of processing the packet comprises the step of inserting a Multiprotocol Label Switching (MPLS) label before an Ethernet header of the packet in accordance with a Pseudowire protocol.

3. The method of claim 1, wherein the step of processing the packet comprises the step of inserting a control word before an Ethernet header of the packet in accordance with a Pseudowire protocol.

4. The method of claim 1, further comprising the steps of:

generating by a second designated processing unit a Generic Associated Channel (G-ACH) message in accordance with a MPLS Transport Profile (MPLS-TP) protocol; and

injecting the G-ACH message into the OpenFlow pipeline for further processing.

5. The method of claim 4, wherein the G-ACH message is an Operations, Administration and Maintenance (OAM) packet.

6. The method of claim 1, wherein each of the processing units has a configurable functionality that is different from each other.

7. The method of claim 1, wherein the processing units are interconnected via configurable communication links to form a hierarchy.

8. The method of claim 1, wherein the processing units are interconnected via configurable communication links to form a chain.

9. The method of claim 1, wherein each of the processing units runs on one or more cores in a multi-core processor.

10. A network element functioning as a switch in a split-architecture network that includes a controller coupled to a set of switches, the switch to provide high speed packet processing and enhanced network functionalities, the network element comprising:

a communication channel coupled to an OpenFlow pipeline that includes a set of flow tables and each of the flow tables including a set of table entries;

a first set of processing cores coupled to the OpenFlow pipeline via the communication channel to perform operations of a set of processing units, each of the processing units operable to process packets with the enhanced network functionalities and to inject the packets back to one of the flow tables in the OpenFlow pipeline, wherein the enhanced network functionalities include functionalities that are not supported by OpenFlow pipeline of the switch and wherein the set of processing units are external to the OpenFlow pipeline;

an ingress port coupled to the OpenFlow pipeline to receive a packet from a network;

a second set of processing cores coupled to the communication channel, the first set of processing cores and the ingress port, to perform operations of the OpenFlow pipeline, the OpenFlow pipeline to match one or more header fields in the packet against the table entries to identify an action in a matching table entry as directing the packet to a designated one of the processing units, and to forward the packet to the designated processing unit for processing; and

an egress port coupled to the OpenFlow pipeline to transmit the packet to the network.

11. The network element of claim 10, wherein the designated processing unit comprises one or more processing cores in the first set of processing cores to insert a Multiprotocol Label Switching (MPLS) label before an Ethernet header of the packet in accordance with a Pseudowire protocol.

12. The network element of claim 10, wherein the designated processing unit comprises one or more processing cores in the first set of processing cores to insert a control word before an Ethernet header of the packet in accordance with a Pseudowire protocol.

13. The network element of claim 10, wherein the set of processing units comprises a second designated processing unit, which comprises one or more processing cores to generate a Generic Associated Channel (G-ACH) message in accordance with a MPLS Transport Profile (MPLS-TP) protocol, and to inject the G-ACH message into the OpenFlow pipeline for further processing.

14. The network element of claim 13, wherein the G-ACH message is an Operations, Administration and Maintenance (OAM) packet.

15. The network element of claim 10, wherein each of the set of processing units has a configurable functionality that is different from each other.

16. The network element of claim 10, wherein the set of processing units are interconnected via configurable communication links to form a hierarchy.

17. The network element of claim 10, wherein the set of processing units are interconnected via configurable communication links to form a chain.

18. The network element of claim 10, wherein the set of processing units are interconnected via configurable mesh connections.