Device, system and/or method for managing packet congestion in a packet switching network

• US 20070201499A1
• Filed: 05/30/2006
• Published: 08/30/2007
• Est. Priority Date: 02/24/2006
• Status: Active Grant

• Current Assignee: Texas Instruments, Inc.
• Original Assignee: Texas Instruments, Inc.
• Inventors: Chang, Fongchi Rex, Kapoor, Pankaj, Manbeck, Jackie Lee
• Granted Patent: US 7,724,754 B2
• Time in Patent Office: Days
• US Class Current: 370/412
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/2408 for supporting different services, e.g. a differentiated services [DiffServ] type of service
• CPC Class Codes: H04L 47/2416 Real-time traffic
• 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 47/30 in combination with information about buffer occupancy at either end or at transit nodes
• 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 47/6215 Individual queue per QOS, rate or priority

Abstract

A network infrastructure device includes a receiver operable to receive packets when operably connected to a communication network; and a processor cooperatively operable with the transceiver. The processor can receive a packet on an interface corresponding to the transceiver. Also, the processor can map the packet to one of several queues and to one of several classifications, based on an indication of priority of handling in a header in the packet and/or an indication of priority in a configuration of the interface. The processor also checks for congestion in the queues with respect to the classification of the packet, and checks for congestion in the one queue with respect to the one classification. The processor queues the packet if there is no congestion, otherwise the processor drops the packet.

Claims

1. A network infrastructure device comprising:

a receiver operable to receive packets when operably connected to a communication network; and

a processor cooperatively operable with the transceiver, the processor being configured to facilitate;

receiving a packet on an interface corresponding to the transceiver;

mapping the packet to one queue of a plurality of queues and to one classification of a plurality of classifications, the mapping being determined on at least one of an indication of priority of handling in a header in the packet and an indication of priority in a configuration of the interface;

first checking for congestion in the plurality of queues with respect to the one classification of the packet and second checking for congestion in the one queue with respect to the one classification; and

queuing the packet if there is no congestion, otherwise dropping the packet.

2. The network infrastructure device of claim 1, wherein the processor is configured to facilitate performing at least the first checking, second checking, queuing and dropping at an ingress side of the processor.

3. The network infrastructure device of claim 1, wherein the processor is further configured to include an operating system, further comprising notifying the operating system that the packet has been received if the packet is queued.

4. The network infrastructure device of claim 1, wherein the communication network is a voice over packet (VOP) network.

5. The network infrastructure device of claim 1, wherein the classifications include a top priority class, and wherein voice packets are mapped to the top priority classification which are not dropped unless all of the ingress queues are congested.

6. The network infrastructure device of claim 1, wherein the processor is further configured to facilitate re-marking the packet, the re-marking including translating the classification mapped to the packet to an equivalent priority of handling in a domain to which the packet is to be transmitted, and indicating the equivalent priority of handling in the header.

7. The network infrastructure device of claim 6, further comprising a transmitter, wherein the processor is further configured to facilitate receiving packets from a receiving domain in accordance with the receiver, and transmitting packets to a domain on which the packet is to be transmitted, in accordance with the transmitter.

8. A computer-implemented method, implemented on a network infrastructure device in a packet switching network, for managing packet congestion, comprising:

at the network infrastructure device, obtaining a packet from an ingress queue of a plurality of ingress queues, where respective ingress queues correspond to respective classifications of a plurality of classifications;

at the network infrastructure device, re-marking the packet obtained from the ingress queue, the re-marking including translating a classification corresponding to the ingress queue to an equivalent priority of handling in a domain to which the packet is to be transmitted, and indicating the equivalent priority of handling in the header;

at the network infrastructure device, placing the packet obtained from the ingress queue into an egress queue responsive to an interface from which the packet is to be transmitted, where there are a plurality of egress queues, including the egress queue, corresponding to each interface of a plurality of interfaces from which packets can be transmitted, where respective egress queues in each of the plurality of egress queues correspond to the respective classifications; and

at the network infrastructure device, selecting the next packet to be transmitted on the interface, from the plurality of egress queues for the interface from which the packet is to be transmitted, responsive to an egress shaper operating on the plurality of egress queues.

9. The method of claim 8, wherein the packet switching network is a voice over packet (VOP) network.

10. The method of claim 8, further comprising transmitting the next packet from the interface.

11. The method of claim 8, wherein the selecting responsive to an egress shaper is performed at an egress side at a layer below an application layer of the network infrastructure device.

12. The method of claim 8, the egress shaper reserving a portion of a bandwidth associated with the interface for at least one particular data type, wherein data types correspond to the respective classifications, and the classification of the packet determines a portion of the bandwidth corresponding to the packet.

13. The method of claim 8, wherein the respective classifications correspond to respective priorities, and the selecting is performed in accordance with the priorities.

14. A computer-readable medium comprising instructions for execution by a computer, the instructions including a computer-implemented method for managing packet congestion in a network infrastructure device operating in a packet switching network, the instructions for implementing:

(A) classifying a plurality of packets at ingress by at least one of an indication of priority of handling in a header in the packets and an indication of priority in a configuration of the interface, dropping the packets if ingress queues are congested, and otherwise placing the packets in the ingress queues, wherein there are a plurality of ingress queues each corresponding to a respective classification;

(B) re-marking the packets in accordance with the classification corresponding to the ingress queues, and placing the packets into the egress queues, wherein there are a plurality of egress queues corresponding to each interface of a plurality of interfaces from which packets can be transmitted, wherein respective egress queues in each of the plurality of egress queues correspond to respective classifications; and

(C) shaping the queues at egress by selecting the next packets to be transmitted on the interfaces from the plurality of egress queues for the interfaces from which the packets are to be transmitted.

15. The computer-readable medium of claim 14, wherein the instructions for classifying and dropping are performed at a layer below an application layer.

16. The computer-readable medium of claim 14, wherein the instructions for shaping are performed below an application layer.

17. The computer-readable medium of claim 14, wherein the re-marking includes instructions for translating the classification to an equivalent priority of handling in a domain to which the packet is to be transmitted, and indicating the equivalent priority of handling in the header.

18. The computer-readable medium of claim 17, wherein the field of the packet that is re-marked include at least one of the user priority field, the type of service field, and the code point field.

19. The computer-readable medium of claim 14, further comprising instructions for reserving a portion of bandwidths associated with the interfaces for at least one particular data type, wherein data types correspond to the respective classifications, and the respective classifications of the packets determines their respective corresponding portion of the bandwidth.

20. The computer-readable medium of claim 14, wherein the packet switching network is a voice over packet (VOP) network and the packets include voice packets, wherein the respective classifications include a top priority class, and wherein voice packets are mapped to the top priority classification and are not dropped unless all of the ingress queues are congested.