Preserving sequencing during selective packet acceleration in a network environment

• US 8,792,353 B1
• Filed: 06/14/2011
• Issued: 07/29/2014
• Est. Priority Date: 06/14/2011
• Status: Active Grant

• Current Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Original Assignee: Cisco Technology, Inc. (Cisco Systems, Inc.)
• Inventors: Janarthanan, Bhagatram Yaugand, Mackie, Robert A., Dixon, Walter, Sunavala, Farhad P.
• Primary Examiner(s): Sheikh, Ayaz
• Assistant Examiner(s): AGA, SORI A
• Application Number:
  US13/160,189
• Time in Patent Office: 1,141 Days
• Field of Search: None
• US Class Current: 370/236
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/2483 involving identification of individual flows
• CPC Class Codes: H04L 49/355 Application aware switches, e.g. for HTTP

Abstract

A first set of frames is received in a data flow and tagged with respective serial numbers identifying an ordering of frames in the first set. The tagged first set is forwarded to a general processor for processing prior to being sent to a remote network element. A data flow policy is received from the general processor based on processing of the first set. A second set of frames of the data flow are received subsequent to forwarding the first set to the general processor. It can be verified, based on the assigned serial numbers, that each frame in the first set has been sent to the remote network element prior to sending frames in the second set. The second set is processed and sent to the remote network element by the network processor, bypassing the general processor, based on the data flow policy.

Claims

1. A method, comprising:

receiving a first set of frames of a first data flow;

tagging each frame in the first set of frames with a respective assigned serial number identifying an ordering of frames in the first set of frames;

forwarding the tagged first set of frames from a network processor to a general processor for processing prior to the first set of frames being sent to at least one remote network element, wherein the first set of frames comprises a plurality of frames;

receiving, at the network processor, following the forwarding of the tagged first set of frames to the general processor, a first data flow policy generated by the general processor based at least in part on processing of at least some frames in the tagged first set of frames, wherein the first data flow policy indicates to the network processor that the first set of frames are to be accelerated using the network processor, and frames in the first data flow received before the first data flow policy are to be first passed to the general processor;

receiving a second set of frames of the first data flow subsequent to forwarding the first set of frames to the general processor and receiving the first data flow policy; and

verifying, based at least in part on the serial numbers assigned to the first set of frames, that each frame in the first set of frames has been sent to the remote network element prior to sending frames in the second set of frames to the remote network element, wherein the second set of frames are to be accelerated by processing and sending at least the second set of frames to the remote network element using the network processor, bypassing processing by the general processor, based at least in part on the first data flow policy.

2. The method of claim 1, wherein verifying that each frame in the first set of frames has been sent to the remote network element includes:

receiving, at the network processor, the tagged first set of frames returned from the general processor following processing of the first set of frames by the general processor;

identifying, using the network processor, the tagged serial number of each frame in the tagged first set of frames returned from the general processor;

sending, using the network processor, the returned first set of frames to the at least one remote network element.

3. The method of claim 1, wherein the first data flow policy includes a request to accelerate at least a portion of the first data flow using the network processor and the second set of frames are processed according to the request to accelerate.

4. The method of claim 3, further comprising sending an acknowledgement from the network processor to the general processor in response to receiving the request to accelerate.

5. The method of claim 1, further comprising:

receiving an intermediate frame of the first data flow after receiving the first set of frames and the first data flow policy and before receiving the second set of frames;

determining, in response to receiving the intermediate frame, that a last received frame in the first set of frames tagged with a last-assigned serial number has not been returned from the general processor and sent to the at least one remote network element;

tagging the intermediate frame with a serial number indicating that the intermediate frame was received after the last-received frame in the first set of frames;

forwarding the tagged intermediate frame to a general processor for processing prior to the intermediate frame being sent to the at least one remote network element.

6. The method of claim 1, wherein the first set of frames includes a last received frame in the first set and the last received frame is assigned a last-assigned serial number, the method further comprising storing, in memory, the value of at least the last-assigned serial number.

7. The method of claim 1, wherein serial numbers assigned to the first set of frames have sequential values, wherein a first frame in the first set of frames, received before a second frame in the first set of frames, is assigned a serial number with a value lower than the serial number assigned to the second frame.

8. The method of claim 1, wherein the first data flow policy is sent with at least one particular frame in the first set of frames returned to the network processor from the general processor.

9. The method of claim 8, wherein the network processor parses the first data flow policy and strips the first data flow policy from the particular frame before the particular frame is sent to the remote network element.

10. The method of claim 1, further comprising stripping tagged serial numbers from the first set of frames prior to sending the first set of frames to the at least one remote network element.

11. The method of claim 1, wherein the first data flow policy includes a set of conditions for bypassing the general processor.

12. The method of claim 11, wherein the set of conditions includes a trigger that, when met, returns processing of at least a portion of the first data flow to the general processor.

13. The method of claim 11, wherein the trigger is a time-based trigger.

14. The method of claim 11, wherein the trigger is a volume-based trigger defining a maximum volume of data to be processed before returning control to the general processor.

15. The method of claim 1, wherein the network processor and general processor are included in a mobile gateway device sending at least a portion of the received first data flow to the remote network element.

16. A system comprising:

at least one memory element for storing data;

at least one general processor configured to;

perform a set of inspection operations on at least one received data flow;

generate an acceleration request for the received data flow, the acceleration request including instructions to accelerate at least a portion of the received data flow by bypassing the general processor; and

at least one network processor configured to;

forward received data flows to at least one remote network element;

tag each frame in a first set of frames in the received data flow with a respective assigned serial number identifying an ordering of frames in the first set of frames, wherein the first set of frames comprises a plurality of frames;

forward the first set of frames, including serial number tags, to the general processor for processing prior to the first set of frames being forwarded to the remote network element, wherein frames in the received data flow received before the acceleration request are to be forwarded to the general processor prior to forwarding to the remote network element;

verify, based at least in part on the serial numbers assigned to the first set of frames, that each frame in the first set of frames has been sent to the remote network element;

process a second set of frames in the received data flow according to the acceleration request, wherein processing of the second set of frames bypasses the general processor and the second set of frames are forwarded to the remote network element after verifying that the first set of frames have been sent to the remote network.

17. The system of claim 16, further comprising:

an in-band communication link between the network processor and the general processor; and

an out-of-band communication link between the network processor and the general processor.

18. The system of claim 17, wherein the network processor tags the first set of frames with the serial number tags and forwards the first set of frames with the serial number tags via the in-band communication channel.

19. The system of claim 17, wherein the network processor forwards the first set of frames via the in-band communication channel and forwards the serial number tags via the out-of-band communication channel.

20. Logic encoded in non-transitory media that includes code for execution and when executed by a processor is operable to perform operations comprising:

receiving, at a network processor, a first set of frames of a first data flow;

tagging each frame in the first set of frames with a respective assigned serial number identifying an ordering of frames in the first set of frames;

forwarding the tagged first set of frames to a general processor for processing prior to the first set of frames being sent to at least one remote network element, wherein the first set of frames comprises a plurality of frames;

receiving, at the network processor, following the forwarding of the tagged first set of frames to the general processor, a first data flow policy generated by the general processor based at least in part on processing of at least some frames in the tagged first set of frames, wherein the first data flow policy indicates to the network processor that the first set of frames are to be accelerated using the network processor, and frames in the first data flow received before the first data flow policy are to be first passed to the general processor;

receiving, at the network processor, a second set of frames of the first data flow subsequent to forwarding the first set of frames to the general processor and receiving the first data flow policy;

verifying, based at least in part on the serial numbers assigned to the first set of frames, that each frame in the first set of frames has been sent to the remote network element prior to sending frames in the second set of frames to the remote network element, wherein the second set of frames are to be accelerated by processing and sending at least the second set of frames to the remote network element using the network processor, bypassing processing by the general processor, based at least in part on the first data flow policy.