APPLICATION-AWARE M:N HOT REDUNDANCY FOR DPI-BASED APPLICATION ENGINES

• US 20100057940A1
• Filed: 08/28/2008
• Published: 03/04/2010
• Est. Priority Date: 08/28/2008
• Status: Active Grant

• Current Assignee: RPX Corporation
• Original Assignee: Alcatel-Lucent SA (Nokia Corporation)
• Inventors: Morin, Steven Edward, Dolganow, Andrew
• Granted Patent: US 8,626,954 B2
• Time in Patent Office: Days
• US Class Current: 709/250
• CPC Class Codes: H04L 43/0817 by checking functioning
• CPC Class Codes: H04L 47/803 Application aware

Abstract

A packet processing system for providing application-aware hot redundancy and a related card and methods are disclosed. The system may include a plurality of active devices, each including a processor configured to perform deep packet inspection to identify an application associated with an active flow, and a redundancy block configured to determine whether the application requires hot redundancy and, if so, to forward a message from which state information may be derived. The system may also include at least one protecting device in communication with each of the active devices and configured to receive the message regarding the active flow from the active device, derive state information from the message, and resume packet forwarding operations for the active flow upon failure of a respective active device of the plurality of active devices.

Claims

1. A packet processing system for providing application-aware hot redundancy, the system comprising:

a plurality of active devices, each of the plurality of active devices comprising.;

a processor configured to perform deep packet inspection to identify an application associated with an active flow, and

a redundancy block configured to determine whether the application requires hot redundancy and, if so, to forward a message from which state information may be derived; and

at least one protecting device providing redundancy for the plurality of active devices, wherein the protecting device is in communication with each of the active devices and is configured to;

receive the message regarding the active flow from a respective active device of the plurality of active devices,

derive state information from the message, and

resume packet forwarding operations for the active flow using the state information upon failure of the respective active device.

2. The packet processing system for providing application-aware hot redundancy according to claim 1, wherein the message forwarded by the redundancy block of the respective active device is a duplicate of a packet belonging to the active flow.

3. The packet processing system for providing application-aware hot redundancy according to claim 1, wherein the message forwarded by the redundancy block of the active device includes state information predetermined by the redundancy block of the respective active device.

4. The packet processing system for providing application-aware hot redundancy according to claim 1, wherein the applications for which hot redundancy is required are specified by a service provider based on at least one of an average duration of flows associated with each application and a criticality of each application.

5. The packet processing system for providing application-aware hot redundancy according to claim 1, wherein the state information includes at least one of an Internet protocol 5-tuple, an application identifier, and application policy information.

6. A card for supporting application-aware hot redundancy in a packet processing system, the card comprising:

a receiving interface configured to receive an incoming packet belonging to an active flow;

a processor configured to perform deep packet inspection to identify an application associated with the active flow; and

a redundancy block configured to;

determine whether the application requires hot redundancy and, if so, to forward a message to a protecting device, the message including sufficient information from which state information is derived by the protecting device, and

communicate with the protecting device, such that the protecting device resumes packet forwarding operations for the active flow using the state information upon failure of the card.

7. The card for supporting application-aware hot redundancy in a packet processing system according to claim 6, wherein the message forwarded by the redundancy block is a duplicate of a packet belonging to the active flow.

8. The card for supporting application-aware hot redundancy in a packet processing system according to claim 6, wherein the message forwarded by the redundancy block includes state information predetermined by the redundancy block.

9. The card for supporting application-aware hot redundancy in a packet processing'"'"'system according to claim 6, wherein the applications for which hot redundancy is required are specified by a service provider based on at least one of an average duration of flows associated with each application and a criticality of each application.

10. A method for providing application-aware hot redundancy in a packet processing system, the method comprising:

receiving redundancy information indicating applications for which hot redundancy is required;

receiving an incoming packet on an active device, the incoming packet belonging to an active flow;

performing deep packet inspection (DPI) to identify an application associated with the active flow;

determining whether hot redundancy is required for the active flow using the identified application and the redundancy information; and

duplicating the incoming packet to a protecting device when it is determined that hot redundancy is required.

11. The method for providing application-aware hot redundancy in a packet processing system according to claim 10, wherein the step of duplicating the incoming packet is executed only when the protecting device requires additional information to identify the application associated with the active flow.

12. The method for providing application-aware hot redundancy in a packet processing system according to claim 10, further comprising:

sending an end of flow indicator to the protecting device when it is determined that hot redundancy is not required for the active flow; and

in response to receipt of the end of flow indicator on the protecting device, performing a clean-up operation on the protecting device to remove information regarding the active flow from memory.

13. The method for providing application-aware hot redundancy in a packet processing system according to claim 10, further comprising:

receiving the packet on the protecting device; and

establishing state information for the active flow on the protecting device when it is determined that the application associated with the active flow is not yet known.

14. The method for providing application-aware hot redundancy in a packet processing system according to claim 13, further comprising:

when the active device fails, utilizing the state information to resume packet forwarding operations for the active flow on the protecting device.

15. The method for providing application-aware hot redundancy in a packet processing system according to claim 10, further comprising:

providing warm redundancy on the protecting device as a best effort service when multiple active devices fail simultaneously, such that the protecting device provides hot redundancy for all flows that require hot redundancy.

16. The method for providing application-aware hot redundancy in a packet processing system according to claim 15, further comprising:

dropping warm redundant flows according to a priority associated with each flow, the priority for each flow determined by considering at least one of a duration of the flow, the application associated with the flow, and a customer associated with the flow.

17. A method for providing application-aware hot redundancy in a packet processing system, the method comprising:

receiving redundancy information indicating applications for which hot redundancy is to be provided;

receiving an incoming packet on an active device, the incoming packet belonging to an active flow;

performing deep packet inspection (DPI) to identify state information associated with the active flow, the state information identifying an application associated with the flow;

determining whether hot redundancy is required for the active flow using the identified application and the redundancy information; and

transmitting a message to a protecting device when it is determined that hot redundancy is required, the message including the state information.

18. The method for providing application-aware hot redundancy in a packet processing system according to claim 17, the method further comprising:

transmitting a last packet indication to the protecting device when it is determined that the incoming packet is the last packet in the active flow; and

in response to receipt of the last packet indication on the protecting device, performing a clean-up operation on the protecting device to remove information regarding the active flow from memory.

19. The method for providing application-aware hot redundancy in a packet processing system according to claim 17, the method further comprising:

receiving the message on the protecting device;

extracting the state information from the message;

storing the state information on the protecting device.

20. The method for providing application-aware hot redundancy in a packet processing system according to claim 19, further comprising:

when the active device fails, utilizing the state information to resume packet forwarding operations for the active flow on the protecting device.

21. The method for providing application-aware hot redundancy in a packet processing system according to claim 17, further comprising:

providing warm redundancy on the protecting device as a best effort service when multiple active devices fail simultaneously, such that the protecting device provides hot redundancy for all flows that require hot redundancy.

22. The method for providing application-aware hot redundancy in a packet processing system according to claim 17, further comprising:

dropping warm redundant flows according to a priority associated with each flow, the priority for each flow determined by considering at least one of a duration of the flow, the application associated with the flow, and a customer associated with the flow.