System and method for latency reduction in a network environment

• US 8,812,725 B2
• Filed: 05/18/2012
• Issued: 08/19/2014
• Est. Priority Date: 05/18/2012
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Kulkarni, Ajay A.
• Primary Examiner(s): JEAN GILLES, JUDE
• Application Number:
  US13/475,522
• Publication Number:
  US 20130311643A1
• Time in Patent Office: 823 Days
• Field of Search: 709/224, 709/229, 709/230, 709/223, 709/232, 709/235
• US Class Current: 709/235
• CPC Class Codes: H04L 41/5022 by giving priorities, e.g. assigning classes of service
• CPC Class Codes: H04L 43/0852 Delays
• CPC Class Codes: H04L 43/0864 Round trip delays
• CPC Class Codes: H04L 43/16 Threshold monitoring
• CPC Class Codes: H04L 47/283 in response to processing delays, e.g. caused by jitter or round trip time [RTT]
• CPC Class Codes: H04L 47/32 by discarding or delaying data units, e.g. packets or frames
• CPC Class Codes: H04L 47/323 Discarding or blocking control packets, e.g. ACK packets
• CPC Class Codes: H04L 69/16 Implementation or adaptation of Internet protocol [IP], of transmission control protocol [TCP] or of user datagram protocol [UDP]

Abstract

A method is provided in one example embodiment that includes identifying a foreground flow that has a higher priority than at least one background traffic flow in a network. The method also includes determining (e.g., periodically) a latency metric associated with the foreground flow, and activating a latency reduction response if the latency metric exceeds a threshold. In more particular embodiments, the latency reduction response may include diverting background flows in the network to a buffer for a delay period and/or delaying acknowledgements for background flows until the latency metric is below the threshold. In yet more particular embodiments, the latency metric may be a round-trip time.

Claims

1. A method, comprising:

identifying a foreground flow that has a higher priority than at least one background traffic flow in a network;

determining a latency metric associated with the foreground flow; and

activating a latency reduction response if the latency metric exceeds a threshold, wherein the latency reduction response comprises diverting one or more background flows in the network to a buffer for a delay period until the latency metric is below the threshold.

2. The method of claim 1, wherein the latency reduction response further comprises delaying acknowledgements for one or more background flows if the latency metric exceeds the threshold.

3. The method of claim 1, wherein the delay period is periodically modified until the latency metric is below the threshold.

4. The method of claim 1, wherein the delay period is periodically modified based on an exponential back-off algorithm until the latency metric is below the threshold.

5. The method of claim 1, wherein identifying the foreground flow comprises identifying inherent characteristics of the foreground flow.

6. The method of claim 1, wherein identifying the foreground flow comprises identifying a flag in a header of the foreground flow.

7. The method of claim 1, wherein the latency metric is a round-trip time.

8. The method of claim 1, wherein the foreground flow is a map-reduce flow.

9. Logic encoded in one or more tangible non-transitory media that includes code for execution and when executed by a processor operable to perform operations comprising:

identifying a foreground flow that has a higher priority than at least one background traffic flow in a network;

determining a latency metric associated with the foreground flow; and

activating a latency reduction response if the latency metric exceeds a threshold, wherein the latency reduction response comprises diverting one or more background flows in the network to a buffer for a delay period until the latency metric is below the threshold.

10. The logic of claim 9, wherein the latency reduction response further comprises delaying acknowledgements for the one or more background flows if the latency metric exceeds the threshold.

11. The logic of claim 9, wherein the delay period is periodically modified until the latency metric is below the threshold.

12. The logic of claim 9, wherein the delay period is periodically modified based on an exponential back-off algorithm until the latency metric is below the threshold.

13. The logic of claim 9, wherein identifying the foreground flow comprises identifying inherent characteristics of the foreground flow.

14. The logic of claim 9, wherein identifying the foreground flow comprises identifying a flag in a header of the foreground flow.

15. The logic of claim 9, wherein the latency metric is a round-trip time.

16. The logic of claim 9, wherein the foreground flow is a map-reduce flow.

17. An apparatus, comprising:

a memory element configured to store data,

a processor operable to execute instructions associated with the data, and

a flow control module, the apparatus being configured to;

identify a foreground flow that has a higher priority than at least one background traffic flow in a network;

determine a latency metric associated with the foreground flow; and

activate a latency reduction response if the latency metric exceeds a threshold, wherein the latency reduction response comprises diverting one or more background flows in the network to a buffer for a delay period until the latency metric is below the threshold.