Switch for dynamically rerouting traffic due to detection of faulty link

• US 7,929,428 B2
• Filed: 11/25/2008
• Issued: 04/19/2011
• Est. Priority Date: 03/03/2000
• Status: Active Grant

• Current Assignee: ADTRAN Incorporated (ADTRAN Holdings, Inc.)
• Original Assignee: ADTRAN Incorporated (ADTRAN Holdings, Inc.)
• Inventors: Fan, Jason C., Bannai, Vinay K., Jogalekar, Prasad P.
• Primary Examiner(s): Kumar; Pankaj
• Assistant Examiner(s): Duong; Duc T
• Application Number:
  US12/323,201
• Publication Number:
  US 20090141621A1
• Time in Patent Office: 875 Days
• Field of Search: 370216-228, 370/242, 370/244, 370/245, 370248-250, 370/400, 370/401, 370/404, 370/405, 709/220, 709/221, 709/239, 709/251
• US Class Current: 370/225
• CPC Class Codes: H04L 2101/622 Layer-2 addresses, e.g. medium access control [MAC] addresses
• CPC Class Codes: H04L 2101/672 Short addresses
• CPC Class Codes: H04L 45/02 Topology update or discovery
• CPC Class Codes: H04L 45/021 Ensuring consistency of routing table updates, e.g. by using epoch numbers
• CPC Class Codes: H04L 61/00 Network arrangements, protocols or services for addressing or naming
• CPC Class Codes: H04L 61/2596 Translation of addresses of the same type other than IP, e.g. translation from MAC to MAC addresses
• CPC Class Codes: H04L 61/35 involving non-standard use of addresses for implementing network functionalities, e.g. coding subscription information within the address or functional addressing, i.e. assigning an address to a function

Abstract

The disclosed network includes two rings, wherein a first ring transmits data in a clockwise direction, and the other ring transmits data in a counterclockwise direction. The traffic is removed from the ring by the destination node. During normal operations (i.e., all spans operational), data between nodes can flow on either ring. Thus, both rings are fully utilized during normal operations. The nodes periodically test the bit error rate of the links (or the error rate is constantly calculated) to detect a fault in one of the links. The detection of such a fault sends a broadcast signal to all nodes to reconfigure a routing table within the node so as to identify the optimum routing of source traffic to the destination node after the fault.

Claims

1. A switch for use in a communications network, said network comprising a plurality of switches interconnected in a ring by communication links, each of said switch comprising:

one or more transceivers for being connected to associated links to one or more other switches;

a switch fabric for routing information to and from said one or more transceivers;

a routing table; and

one or more processors, said one or more processors for controlling said switch to;

test the quality of links between switches, wherein said one or more processors control said switch to periodically transmit and receive test messages to and from neighboring switches in said ring and detect the quality of links carrying said test messages;

detect that one or more first links, when connected to said switch, do not meet a quality threshold;

transmit information from said switch to other switches to identify said one or more first links;

determine reroute paths for incoming traffic to said switch, such that the rerouted traffic does not traverse said one or more first links, due to said one or more first links being faulty, wherein each of said plurality of switches in the communications network determines said reroute paths independently of the other switches in the communications network;

revise said routing table to reroute traffic according to the reroute paths; and

route traffic to a destination switch, based on a revised routing table, so as to route said traffic in a direction around said ring different from a direction that the traffic would have traveled to said destination switch had said one or more first links not been faulty.

2. The routing switch of claim 1 wherein said one or more processors includes a CPU connected to said switch fabric.

3. The routing switch of claim 1 wherein said one or more processors control said routing switch to detect the quality of said links based on comparing a bit error rate to a threshold.

4. The routing switch of claim 1 wherein said one or more processors control said routing switch to detect the quality of said links by detecting a loss of an optical signal.

5. The switch of claim 1 wherein said one or more processors control said switch to detect the quality of said links by detecting a loss of an electrical signal.

6. The switch of claim 1 wherein said one or more processors control said switch to detect the quality of said links by detecting a loss of a frame.

7. The switch of claim 1 wherein said one or more processors control said switch to receive messages from other switches in said ring identifying a faulty link and revise said routing table to reroute traffic.

8. The switch of claim 1 wherein said one or more processors control said switch to periodically transmit link status messages along with a session number, create a new session number when it has been detected that said one or more first links do not meet said quality threshold, and transmit the identity of any faulty links along with said new session number to other switches in said ring.

9. The switch of claim 8 wherein said one or more processors control said switch to compare a transmitted session number to a stored session number and, if the session number is different, revise said routing table to take into account a faulty link.

10. The switch of claim 9 wherein said one or more processors control said switch to revise said routing table to identify optimum routes for traffic destinations.

11. The switch of claim 10 wherein said optimum routes are a minimum number of hops to a destination.

12. The switch of claim 10 wherein said optimum routes are lowest cost routes.

13. The switch of claim 12 wherein said lowest cost routes have the lowest aggregate bit error rates.

14. The switch of claim 12 wherein said lowest cost routes are based on a level of traffic congestion.

15. The switch of claim 1 wherein said one or more processors for controlling said switch to route traffic to a destination node control said switch to append a routing header to a message to be sent to said destination node.

16. The switch of claim 1 wherein said traffic comprises packets.

17. The switch of claim 1 wherein said traffic comprises cells.

18. The switch of claim 1 wherein said one or more processors for controlling said switch to route traffic to a destination node control said switch to drop certain types of traffic due to a reduced bandwidth in said ring when said one or more first links are faulty.

19. The switch of claim 1 wherein said one or more processors for controlling said switch to revise said routing table to reroute traffic control said switch to wrap inbound traffic headed for a switch with a faulty ink around a different direction in said ring.

20. The switch of claim 1 wherein said one or more processors for controlling said switch to transmit information from said switch to other switches to identify said one or more first links control said switch to require an acknowledgement by said other switches that a link status message has been received and, if said acknowledgement has not been received, re-transmit said status message.

21. The switch of claim 1 wherein the said traffic is processed by said switch at a rate greater than 1 gigabit per second.

22. The switch of claim 1 wherein said one or more processors control said switch to detect that one or more first links, when connected to said switch, meet a quality threshold, after not meeting said quality threshold, and transmit information to other switches to identify that said one or more first links now meet said quality threshold.

23. The switch of claim 1 wherein said one or more processors for controlling said switch to transmit information from said switch to other switches in a direction opposite to the network along the ring to identify said links.

24. The switch of claim 1 wherein said one or more processors for controlling said switch to change a direction of network traffic bound for an egress link corresponding to said links to an opposite direction along the ring.

25. The switch of claim 1 wherein said one or more processors for controlling said switch to determine said reroute paths for said traffic by said switch, such that a direction of the traffic is in an opposite direction along the ring.

26. The switch of claim 1 wherein said one or more processors for controlling said switch to determine if the traffic is protected traffic or unprotected traffic, and to route the traffic to the destination switch, such that protected traffic is less affected by said links than the unprotected traffic.

27. A method for rerouting traffic in a communications network, the communications network comprising a plurality of links between a plurality of switches, comprising the steps of:

(a) detecting a faulty link by one of the plurality of switches, wherein the plurality of switches are interconnected in a ring;

(b) transmitting information from the one switch to the remaining of the plurality of switches to identify the faulty link;

(c) determining reroute paths for the traffic by the plurality of switches, such that the rerouted traffic does not traverse the faulty link, wherein each of the plurality of switches determines the reroute paths independently of the other switches;

(d) revising a routing table at each of the plurality of switches according to the reroute paths; and

(e) routing the traffic to a destination switch based on the revised routing tables of the plurality of switches, wherein routing the traffic to a destination switch comprises,

(e1) determining if the traffic is protected traffic or unprotected traffic;

(e2) routing the traffic to the destination switch, such that protected traffic is less affected by the faulty link than the unprotected traffic.

28. The method of claim 27, wherein the faulty link comprises an ingress link to the one switch.

29. The method of claim 27, wherein the transmitting step (b) comprises:

(b1) sending a notification message from the one switch along the ring, in a direction opposite to the network along the ring, to the remaining of the plurality of switches to identify the faulty link.

30. The method of claim 27, wherein the determining step (c) comprises:

(c1) changing a direction of the network traffic bound for an egress link corresponding to the faulty link to an opposite direction along the ring; and

(c2) determining reroute paths for the traffic by the plurality of switches, such that the rerouted traffic does not traverse the faulty link, wherein each of the plurality of switches independently determines the reroute paths.

31. The method of claim 27, wherein the revising step (d) comprises:

(d1) determining the reroute paths for the traffic by each of the plurality of switches, such that a direction of the traffic is in an opposite direction along the ring.

32. A non-transitory computer readable medium with program instructions for rerouting traffic in a communications network, the communications network comprising a plurality of links between a plurality of switches, comprising the instructions for:

(a) detecting a faulty link by one of the plurality of switches, wherein the plurality of switches are interconnected in a ring;

(b) transmitting information from the one switch to the remaining of the plurality of switches to identify the faulty link; and

(c) determining reroute paths for incoming traffic to the plurality of switches by the plurality of switches, such that the rerouted traffic does not traverse the faulty link, wherein each of the plurality of switches determines the reroute paths independently of the other switches;

(d) revising a routing table at each of the plurality of switches according to the reroute paths; and

(e) routing the traffic to a destination switch based on the revised routing tables of the plurality of switches, wherein routing the traffic to a destination switch comprises,

(e1) determining if the traffic is protected traffic or unprotected traffic;

(e2) routing the traffic to the destination switch, such that protected traffic is less affected by the faulty link than the unprotected traffic.