Dynamic latency-based rerouting

• US 9,154,394 B2
• Filed: 09/28/2010
• Issued: 10/06/2015
• Est. Priority Date: 09/28/2010
• Status: Expired due to Fees

• Current Assignee: Avago Technologies International Sales Pte Limited (Broadcom, Inc.)
• Original Assignee: Brocade Communications Systems, Inc. (Broadcom, Inc.)
• Inventors: Ko, Kung-Ling, Varanasi, Surya Prakash, Altekar, Satsheel B., Terry, John Michael, Balakavi, Vankata Pramod
• Primary Examiner(s): Nickerson, Jeffrey
• Assistant Examiner(s): PATEL, JANAKKUMAR T
• Application Number:
  US12/892,115
• Publication Number:
  US 20120075999A1
• Time in Patent Office: 1,834 Days
• Field of Search: 370238-239
• US Class Current: 1/1
• CPC Class Codes: H04L 1/00 Arrangements for detecting or preventing errors in the information received correcting synchronisation H04L7/00
• CPC Class Codes: H04L 43/0858 One way delays
• CPC Class Codes: H04L 45/121 by minimising delays
• CPC Class Codes: H04L 45/28 using route fault recovery
• CPC Class Codes: H04L 45/70 Routing based on monitoring results
• CPC Class Codes: H04L 47/12 Avoiding congestion; Recovering from congestion

Abstract

A switch creates and dynamically updates a latency map of a network to adjust routing of flows. Further, the network is monitored to detect latency issues and trigger a dynamic adjustment of routing based on the latency map. In this manner, a flow can be routed along a route (i.e., a faster route) that provides less latency than other available routes. The latency map can be generated based on latency probe packets that are issued from and returned to the source switch. By evaluating many such latent probe packets that have traveled along many available routes (e.g., corresponding to various ports of the switch), the switch or associated administrative logic can dynamically adjust the latency map to updated latency information of available routes. Therefore, responsive to a trigger, the source switch can dynamically adjust the routing of a flow based on latency issues discerned from the network.

Claims

1. A switch, comprising:

a latency management module including a processor configured to identify a probe packet from a remote switch identifying latency associated with one of at least two routes between the switch and the remote switch;

a packet processor adapted to;

remove the probe packet from a head of a first transmission queue of a first port of the switch, wherein the first port is an egress port associated with a destination device;

in response to the removal of the probe packet, modify the probe packet by;

updating a destination identifier of the probe packet with a source identifier of the probe packet; and

in response to determining that the probe packet is on an outgoing portion of a round trip, updating a value in the probe packet to indicate that the probe packet is on a return portion of the round trip; and

insert the modified probe packet to a head of a second transmission queue of a second port of the switch, wherein the second port is an egress port associated with the remote switch.

2. The switch of claim 1, wherein the switch is one of:

a Fibre Channel (FC) switch;

an Ethernet switch;

a routing bridge (RBridge);

an IP router; and

an multi-protocol label switching (MPLS) switch.

3. The switch of claim 1, wherein respective latency information associated with a respective one of the at least two routes is stored in a data structure at the remote switch.

4. The switch of claim 1, wherein a route is specified by at least one of:

a combination of an identifier of the switch and an identifier of a port on the switch;

a combination of an identifier of the switch and an identifier of an output port of the remote switch; and

a combination of an identifier of the switch and an identifier of a next-hop switch of the remote switch.

5. The switch of claim 1, wherein ingress port of the switch for a flow is changed in response to latency associated with the one of at least two routes being above a predetermined threshold.

6. The switch of claim 5, wherein while changing the ingress port for the flow, transmission of the flow is temporarily suspended.

7. The switch of claim 5, wherein the switch is an egress switch or an intermediary switch for forwarding the flow.

8. A method, comprising:

identifying a probe packet from a second identifying latency associated with one of at least two routes between a first switch and the second switch;

removing the probe packet from a head of a first transmission queue of a first port of the first switch, wherein the first port is an egress port associated with a destination device;

in response to the removal of the probe packet, modifying the probe packet by;

updating a destination identifier of the probe packet with a source identifier of the probe packet; and

in response to determining that the probe packet is on an outgoing portion of a round trip, updating a value in the probe packet to indicate that the probe packet is on a return portion of the round trip; and

inserting the modified probe packet to a head of a second transmission queue of a second port of the first switch, wherein the second port is an egress port associated with the second switch.

9. The method of claim 8, wherein the first switch is one of:

a Fibre Channel (FC) switch;

an Ethernet switch;

a routing bridge (RBridge);

an IP router; and

an multi-protocol label switching (MPLS) switch.

10. The method of claim 8, wherein respective latency information associated with a respective one of the at least two routes is stored in a data structure at the second switch.

11. The method of claim 8, wherein a route is specified by at least one of:

a combination of an identifier of the first switch and an identifier of a port on the first switch;

a combination of an identifier of the first switch and an identifier of an output port of the second switch; and

a combination of an identifier of the first switch and an identifier of a next-hop switch of the second switch.

12. The method of claim 8, wherein ingress port of the first switch for a flow is changed in response to latency associated with the one of at least two routes being above a predetermined threshold.

13. The method of claim 12, wherein changing the ingress port for the flow, transmission of the flow is temporarily suspended.

14. The method of claim 12, wherein the first switch is an egress switch or an intermediary switch for forwarding the flow.

15. A network, comprising:

a first switch; and

a second switch;

wherein the first switch comprises;

a latency probing module adapted to generate a latency probe packet identifying latency associated with one of at least two routes between the first switch and the second switch;

a latency management module adapted to dynamically determine a route for a flow between the first switch and the second switch based on the identified latency; and

wherein the second switch comprises;

a latency management module adapted to identify the probe packet from the first switch identifying latency associated with one of at least two routes between the first switch and the second switch; and

a packet processor adapted to;

remove the probe packet from a head of a first transmission queue of a first port of the second switch, wherein the first port is an egress port associated with a destination device;

in response to the removal of the probe packet, modify the probe packet by;

updating a destination identifier of the probe packet with a source identifier of the probe packet; and

in response to determining that the probe packet is on an outgoing portion of a round trip, updating a value in the probe packet to indicate that the probe packet is on a return portion of the round trip; and

insert the modified probe packet to a head of a second transmission queue of a second port of the second switch, wherein the second port is an egress port associated with the first switch.

16. The network of claim 15, wherein the first or second switch is one of:

a Fibre Channel (FC) switch;

an Ethernet switch;

a routing bridge (RBridge);

an IP router; and

an multi-protocol label switching (MPLS) switch.

17. The network of claim 15, wherein the first switch further comprises a data structure which stores respective latency information associated with a respective one of the at least two routes.

18. The network of claim 15, wherein a route is specified by at least one of:

a combination of an identifier of the second switch and an identifier of a port on the second switch;

a combination of an identifier of the second switch and an identifier of an output port of the first switch; and

a combination of an identifier of the second switch and an identifier of a next-hop switch of the first switch.

19. The network of claim 15, wherein the latency management module of the first switch is further configured to change egress port of the first switch for a flow in response to determining the latency associated with a route to be above a predetermined threshold.

20. The network of claim 19, wherein while changing the egress port for the flow, transmission of the flow is temporarily suspended.

21. The network of claim 15, wherein the first switch is an ingress switch or an intermediary switch for forwarding the flow.