Method and apparatus for discarding data packets

• US 6,647,424 B1
• Filed: 05/19/1999
• Issued: 11/11/2003
• Est. Priority Date: 05/20/1998
• Status: Expired due to Term

• Current Assignee: RPX Clearinghouse LLC (RPX Corporation)
• Original Assignee: Nortel Networks Limited (Nortel Networks Corporation)
• Inventors: Cabeca, Linda, Kidder, Joe, Manor, Larry, Colley, Tom, Pearson, Terry
• Primary Examiner(s): Burgess, Glenton B.
• Assistant Examiner(s): Flynn, Kimberly
• Application Number:
  US09/314,563
• Time in Patent Office: 1,637 Days
• Field of Search: 370/397, 370/412, 370/235, 370/236, 370/321, 370/229, 370/419, 370/337, 709/232, 709/235
• US Class Current: 709/232
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/20 Traffic policing
• CPC Class Codes: H04L 47/2408 for supporting different services, e.g. a differentiated services [DiffServ] type of service
• CPC Class Codes: H04L 47/2441 relying on flow classification, e.g. using integrated services [IntServ]
• CPC Class Codes: H04L 47/2458 Modification of priorities while in transit
• CPC Class Codes: H04L 47/2491 Mapping quality of service [QoS] requirements between different networks
• CPC Class Codes: H04L 47/29 using a combination of thresholds
• CPC Class Codes: H04L 47/31 by tagging of packets, e.g. using discard eligibility [DE] bits
• CPC Class Codes: H04L 47/32 by discarding or delaying data units, e.g. packets or frames
• CPC Class Codes: H04L 49/90 Buffering arrangements
• CPC Class Codes: H04L 9/40 Network security protocols

Abstract

A method, apparatus and article of manufacture for filtering a data packet. The data packet is classified a filter type according to a classification criteria. A size value of the data packet is added to a data count for the filter type. The filter type is compared to a committed information rate (CIR). The data packet is filtered if the data count exceeds the CIR or a burst rate. A status the data packet may optionally be altered if the data count exceeds the CIR. A priority of the data packet may optionally be reclassified if the data count exceeds the CIR.

Claims

1. A method for filtering a data packet comprising the steps of:

classifying an internal service class (ISC) to the data packet according to a classification criteria selected from the group comprising at least two of an ingress channel identifier, an IP source address, an IP destination address, a protocol, an options field, a type of service (TOS)/precedence field, source ports, destination ports, and TCP flags;

adding a size value of the data packet to a data count for the ISC;

comparing the ISC to a committed information rate (CIR);

setting a drop preference bit (DP) for the data packet if the data count exceeds the CIR;

setting a watermark indicator (WM) for the data packet responsive to a congestion level of a buffer memory;

concatenating the ISC, DP, and WM into a key for a lookup into a configurable congestion clip table; and

discarding the data packet responsive to the lookup into the configurable congestion clip table.

2. The method of claim 1 further comprising discarding the data packet if the data count exceeds a burst rate being greater than the CIR.

3. The method of claim 1 further comprising altering the ISC of the data packet if the data count exceeds the CIR.

4. A method for filtering a data packet comprising the steps of:

classifying an internal service class (ISC) selected from the group comprising a sender, an application, a service class, and a destination;

adding a size value of the data packet to a data count for the ISC;

comparing the ISC to a committed information rate (CIR);

setting a drop preference bit (DP) for the data packet if the data count exceeds the CIR;

setting a watermark indicator (WM) for the data packet responsive to a congestion level of a buffer memory;

concatenating the ISC, DP, and WM into a key for a lookup into a configurable congestion clip table; and

discarding the data packet responsive to the lookup into the configurable congestion table.

5. The method of claim 4 further comprising discarding the data packet if the data count exceeds a burst rate being greater than the CIR.

6. The method of claim 4 further comprising altering the ISC of the data packet if the data count exceeds the CIR.

7. A method for filtering a data packet comprising:

classifying an internal service class (ISC) to the data packet according to a classification criteria;

adding a size value of the data packet to a data count for the ISC;

comparing the ISC to a committed information rate (CIR);

setting a drop preference bit (DP) for the data packet if the data count exceeds the CIR;

setting a watermark indicator (WM) for the data packet responsive to a congestion level of a buffer memory;

concatenating the ISC, DP, and WM into a key for a lookup into a configurable congestion clip table;

discarding the data packet responsive to the lookup into the configurable congestion clip table; and

initializing the data count at a time interval.

8. An apparatus for filtering a data packet comprising:

a classifier to classify an internal service class (ISC) to the data packet according to a classification criteria selected from the group comprising an ingress channel identifier, an IP source address, an IP destination address, a protocol, an options field, a TOS/precedence field, source ports, destination ports, and TCP flags;

an adder logically coupled to the classifier to add a size value of the data packet to a data count for the ISC;

a comparator logically coupled to the adder to compare the ISC to a CIR;

a drop flagger logically coupled to the comparator to set a drop preference bit (DP) for the packet if the data count exceeds the CIR;

a watermaker logically coupled to a buffer memory to set a watermark indicator (WM) for the data packet responsive to a congestion level of the buffer memory;

a table lookup unit logically coupled to a configurable congestion clip table to concatenate the ISC, DP, and WM into a key and to lookup the key in the congestion clip table; and

a filter logically coupled to the table lookup unit to discard the data packet responsive to the lookup into the configurable congestion clip table.

9. The apparatus of claim 8 wherein the filter is further to discard the data packet if the data count exceeds a burst rate.

10. The apparatus of claim 8 further comprising a reclassifier to reclassify the ISC of the data packet if the data count exceeds the CIR.

11. An apparatus for filtering a data packet comprising:

a classifier to classify an internal service class (ISC) selected from the group comprising a sender, an application, a service class, and a destination;

an adder logically coupled to the classifier to add a size value of the data packet to a data count for the ISC;

a comparator logically coupled to the adder to compare the ISC to a CIR;

a drop flagger logically coupled to the comparator to set a drop preference bit (DP) for the packet if the data count exceeds the CIR;

a watermaker logically coupled to a buffer memory to set a watermark indicator (WM) for the data packet responsive to a congestion level of the buffer memory;

a table lookup unit logically coupled to the data packet to concatenate the ISC, DP, and WM into a key and to lookup the key in a configurable congestion clip table; and

a filter logically coupled to the table lookup unit to discard the data packet responsive to the lookup into the configurable congestion clip table.

12. The apparatus of claim 11 wherein the filter is further to discard the data packet if the data count exceeds a burst rate.

13. The apparatus of claim 11 further comprising a reclassifier to reclassify the ISC of the data packet if the data count exceeds the CIR.

14. An apparatus comprising:

a classifier to classify an internal service class (ISC) to the data packet according to a classification criteria;

an adder logically coupled to the classifier to add a size value of the data packet to a data count for the ISC;

a comparator logically coupled to the adder to compare the ISC to a CIR;

a drop flagger logically coupled to the comparator to set a drop preference bit (DP) for the packet if the data count exceeds the CIR;

a watermaker logically coupled to a buffer memory to set a watermark indicator (WM) for the data packet responsive to a congestion level of the buffer memory;

a table lookup unit logically coupled to a configurable congestion clip table to concatenate the ISC, DP, and WM into a key and to lookup the key in the congestion clip table;

a filter logically coupled to the table lookup unit to discard the data packet responsive to the lookup into the configurable congestion clip table; and

an initalizer logically coupled to the adder to initalize the data count at a time interval.

15. An article of manufacture for use in a computer system to filter a data packet, the article of manufacture comprising:

computer usable medium having computer readable program code means embodied in the medium, the program code means including;

computer readable program code means embodied in the computer usable medium for causing a computer to initialize the data count at a time interval;

computer readable program code means embodied in the computer usable medium for causing a computer to classify an internal service class (ISC) to the data packet according to a classification criteria;

computer readable program code means embodied in the computer usable medium for causing a computer to add a size value of the data packet to a data count for the ISC;

computer readable program code means embodied in the computer usable medium for causing a computer to compare the ISC to a CIR;

computer readable program code means embodied in the computer usable medium for causing a computer to set a drop preference bit (DP) for the data packet if the data count exceeds the CIR;

computer readable program code means embodied in the computer usable medium for causing a computer to set a watermark indicator (WM) for the data packet responsive to a congestion level of the buffer memory;

computer readable program code means embodied in the computer usable medium for causing a computer to concatenate the ISC, DP, and WM into a key and to lookup the key in a configurable congestion clip table; and

computer readable program code means embodied in the computer useable medium for causing a computer to discard the data packet responsive to the lookup into the configurable congestion clip table.

16. The article of manufacture of claim 15 wherein the computer readable program code means embodied in the computer usable medium for causing the computer to classify the ISC to the data packet according to the classification criteria comprises classifying the ISC to the data packet according to a classification criteria selected from the group comprising ingress channel identifier, IP source address, IP destination address, protocol, options field, TOS/precedence field, source ports, destination ports, and TCP flags.

17. The article of manufacture of claim 15 wherein the computer readable program code means embodied in the computer usable medium for causing the computer to classify the ISC to the data packet according to a classification criteria comprises classifying the ISC selected from the group comprising sender, application, service class, and destination.

18. The article of manufacture of claim 15 wherein the computer readable program code means embodied in the computer usable medium for causing the computer to discard the data packet responsive to the lookup into the configurable congestion clip table is further causing the computer to discard the data packet if the data count exceeds a burst rate.

19. The article of manufacture of claim 15 wherein the computer readable program code means embodied in the computer usable medium further comprises computer readable program code means embodied in the computer usable medium for causing a computer to reclassify the ISC of the data packet if the data count exceeds the CIR.