Traffic switching using multi-dimensional packet classification

• US 7,260,102 B2
• Filed: 02/22/2002
• Issued: 08/21/2007
• Est. Priority Date: 02/22/2002
• Status: Active Grant

• Current Assignee: Ciena Corporation
• Original Assignee: Nortel Networks Limited (Nortel Networks Corporation)
• Inventors: Smith, Roland A., Wight, Mark, Fan, Xiaoming, Mehrvar, Hamid R.
• Primary Examiner(s): Chan; Wing
• Assistant Examiner(s): HALIYUR, VENKATESH N
• Application Number:
  US10/079,559
• Publication Number:
  US 20030161303A1
• Time in Patent Office: 2,006 Days
• Field of Search: 370/389, 370/390, 370/392, 370/401, 370/412, 370/428, 370/432, 370/400
• US Class Current: 370/412
• CPC Class Codes: H04L 2012/5651 Priority, marking, classes
• CPC Class Codes: H04L 47/24 Traffic characterised by specific attributes, e.g. priority or QoS
• CPC Class Codes: H04L 47/2433 Allocation of priorities to traffic types
• CPC Class Codes: H04L 47/2441 relying on flow classification, e.g. using integrated services [IntServ]
• CPC Class Codes: H04L 47/50 Queue scheduling
• CPC Class Codes: H04L 49/1515 Non-blocking multistage, e.g. Clos
• CPC Class Codes: H04L 49/153 ATM switching fabrics having parallel switch planes
• CPC Class Codes: H04L 49/20 Support for services
• CPC Class Codes: H04L 49/25 Routing or path finding in a switch fabric
• CPC Class Codes: H04L 49/30 Peripheral units, e.g. input or output ports
• CPC Class Codes: H04L 49/3081 ATM peripheral units, e.g. policing, insertion or extraction

Abstract

A method and system for conveying an arbitrary mixture of high and low latency traffic streams across a common switch fabric implements a multi-dimensional traffic classification scheme, in which multiple orthogonal traffic classification methods are successively implemented for each traffic stream traversing the system. At least two diverse paths are mapped through the switch fabric, each path being optimized to satisfy respective different latency requirements. A latency classifier is adapted to route each traffic stream to a selected path optimized to satisfy latency requirements most closely matching a respective latency requirement of the traffic stream. A prioritization classifier independently prioritizes traffic streams in each path. A fairness classifier at an egress of each path can be used to enforce fairness between responsive and non-responsive traffic streams in each path. This arrangement enables traffic streams having similar latency requirements to traverse the system through a path optimized for those latency requirements.

Claims

1. A system for conveying an arbitrary mixture of high and low latency traffic streams across a common switch fabric, the system comprising:

at least two diverse paths mapped though the switch fabric from a common input interface to a common output interface, each path being optimized to satisfy respective different traffic latency requirements;

a latency classifier adapted to route each one of the traffic streams received at the input interface to one of the at least two diverse paths based upon a latency requirement of each traffic stream most closely matching the respective traffic latency of each of the at least two diverse paths;

at least two prioritization classifiers, each one of the prioritization classifiers associated with one of the at least two diverse paths, each prioritization classifier independently prioritizes traffic being conveyed through the respective path, and

wherein each one of the traffic streams received at the common input interface is routed to one of the at least two diverse paths by the latency classifier and each of the at least two diverse paths are processed independently by the respective prioritization classifiers before transport through the switch fabric to the common output interface.

2. A system as claimed in claim 1, wherein each path is mapped through respective different physical infrastructure of the switch fabric.

3. A system as claimed in claim 1, wherein two or more paths are mapped through a common physical infrastructure capable of supporting the path optimized to satisfy the most demanding latency requirements.

4. A system as claimed in claim 1, wherein the latency classifier is adapted to selectively couple each one of a plurality of upstream channels of the network to a selected one of the paths, such that a respective traffic stream of a communications session mapped through one of the upstream channels is automatically routed to the selected path.

5. A system as claimed in claim 4, wherein the selected path is determined at a time of set-up of the communications session.

6. A system as claimed in claim 4, wherein the latency classifier is further adapted to extract network management and critical mission message content from each traffic stream.

7. A system as claimed in claim 6, wherein the latency classifier is further adapted to route the extracted network management and critical mission message content to a different path from that of each traffic stream.

8. A system as claimed in claim 1, wherein each path comprises a respective input queue adapted to buffer data of each traffic stream being conveyed through the path.

9. A system as claimed in claim 1, wherein the input queue comprises:

at least two parallel buffers, each buffer being adapted to store data of at least one traffic stream being conveyed through the path; and

a scheduler for controlling transmission of data from each buffer through the path.

10. A system as claimed in claim 9, wherein the traffic streams being conveyed through the path comprise network management and critical mission messages, and the prioritization classifier is adapted to route the network management and critical mission messages to a selected one of the parallel buffers.

11. A system as claimed in claim 9, wherein the prioritization classifier is adapted to route each traffic stream to a selected one of the buffers based on a content of a predetermined field of the respective overhead of each traffic stream.

12. A system as claimed in claim 11, wherein the predetermined field comprises a respective K-byte field of each Synchronous Optical Network (SONET) traffic stream being conveyed through the path.

13. A system as claimed in claim 11, wherein the predetermined field comprises a respective Differentiated Services Code Point (DSCP) field of each Internet Protocol (IP) traffic stream being conveyed through the path.

14. A system as claimed in claim 1, further comprising a fairness classifier disposed at an egress of each path, each fairness classifier being adapted to separate responsive and non-responsive traffic streams received through the respective path.

15. A system as claimed in claim 14, wherein the fairness classifier is adapted to route each of the responsive and non-responsive traffic streams to a respective buffer.

16. A method of conveying an arbitrary mixture of high and low latency traffic streams across a common switch fabric, the method comprising a the steps of:

mapping at least two diverse paths through the switch fabric from a common input interface to a common output interface, each path being optimized to satisfy respective different traffic latency requirements;

routing each traffic stream received at the common input interface to a selected one of the at least two diverse paths, the selected path being optimized to satisfy latency requirements most closely matching a respective latency of the traffic stream; and

processing traffic in each of the at least two diverse paths by independently prioritizing the traffic to be conveyed through the switch fabric by each respective path to a common output interface.

17. A method as claimed in claim 16, further comprising a step of buffering each traffic stream being conveyed through the path within a respective input queue of the path.

18. A method as claimed in claim 16, further comprising a step of selectively coupling each one of a plurality of upstream channels to a selected one of the paths, such that a respective traffic stream of a communications session mapped through one of the upstream channels is automatically routed to the selected path.

19. A method as claimed in claim 18, wherein the selected path is determined at a time of set-up of the communications session.

20. A method as claimed in claim 18, further comprising steps of:

extracting network management and critical mission message content from each traffic stream; and

routing the extracted message content to a different path from that of each traffic stream.

21. A method as claimed in claim 16, wherein the step of controlling the priority of each traffic stream further comprises the steps of:

routing each traffic stream to a selected one of at least two parallel buffers, each buffer being adapted to store data of at least one traffic stream being conveyed through the path, the selected buffer being selected based on a content of a predetermined field of the respective overhead of each traffic stream; and

controlling transmission of data from each buffer through the path.

22. A method as claimed in claim 21, wherein the predetermined field comprises a respective K-byte field of each Synchronous Optical Network (SONET) traffic stream being conveyed through the path.

23. A method as claimed in claim 21, wherein the predetermined field comprises a respective Differentiated Services Code Point (DSCP) field of each Internet Protocol (IP) traffic stream being conveyed through the path.

24. A method as claimed in claim 16, further comprising a step of separating responsive and non-responsive traffic streams at a respective egress end of each path.