Control of Data Flow in a Network

• US 20080304413A1
• Filed: 01/31/2006
• Published: 12/11/2008
• Est. Priority Date: 01/31/2005
• Status: Active Grant

• Current Assignee: British Telecommunications PLC (BT Group PLC)
• Original Assignee: British Telecommunications PLC (BT Group PLC)
• Inventors: Briscoe, Robert J., Jacquet, Arnaud
• Granted Patent: US 8,498,210 B2
• Time in Patent Office: Days
• US Class Current: 370/235
• CPC Class Codes: H04L 69/16 Implementation or adaptation of Internet protocol [IP], of transmission control protocol [TCP] or of user datagram protocol [UDP]
• CPC Class Codes: H04L 69/161 Implementation details of TCP/IP or UDP/IP stack architecture; Specification of modified or new header fields
• CPC Class Codes: H04L 69/163 In-band adaptation of TCP data exchange; In-band control procedures

Abstract

Methods and apparatus for encoding a number of small dynamic values over an n-bit field. Also methods and apparatus allowing for the stateless extraction of separate sequences from repetitions of a single field whereby to communicate more than one signal at once.

Claims

1. A method of encoding information using a plurality of messages traversing a plurality of nodes in a network, said messages each having a header associated therewith comprising one or more fields, at least a selected portion of each header providing at least three codepoints;

said method comprising assigning a codepoint from a first set of codepoints of the selected portion in headers associated with a first proportion of messages, and assigning a codepoint from a second set of codepoints of the selected portion in headers associated with a second proportion of messages, characterised in that

said first proportion is determined in dependence on a first network characteristic;

said second proportion is determined in dependence on a second network characteristic; and

at least one of said first and second network characteristics is variable independently of the other of said network characteristics.

2. A method according to claim 1 wherein said selected portion is an n-bit field, where n is greater than 1.

3. A method according to claim 2 wherein said n-bit field is a 2-bit field.

4. A method according to claim 1, wherein said selected portion comprises an existing field in the header associated with each message.

5. A method according to claim 1, wherein said selected portion comprises a contiguous subset of bits selected from the header associated with each message.

6. A method according to claim 1, wherein said selected portion comprises a non-contiguous subset of bits selected from the header associated with each message.

7. A method according to claim 1, wherein said selected portion comprises a Congestion Notification field.

8. A method according to claim 1, wherein said selected portion comprises the Explicit Congestion Notification field.

9. A method according to claim 1, wherein said information comprises at least one value.

10. A method according to claim 1, wherein at least one of said network characteristics is a dynamic characteristic.

11. A method according to claim 1, wherein said first and second network characteristics are dynamic characteristics.

12. A method according to claim 1, wherein at least one of said network characteristics relates to congestion in said network.

13. A method according to claim 1, wherein at least one of said network characteristics relates to a measure of end-to-end congestion on a path in said network.

14. A method according to claim 13, wherein said network characteristic relating to end-to-end path congestion is dynamic in the sense that it may vary after a period corresponding to a round-trip time of a message traversing said path.

15. A method according to claim 1, wherein at least one of said network characteristics relates to a measure of upstream congestion on a path in said network.

16. A method according to claim 15, wherein said network characteristic relating to upstream congestion is dynamic in the sense that it may vary during a period corresponding to a round-trip time of a message traversing said path.

17. A method according to claim 15, wherein said selected portion comprises the Explicit Congestion Notification field, and wherein a codepoint CE indicative of “

Congestion Experienced”

is assigned to the Explicit Congestion Notification field in the headers associated with a first proportion of messages in dependence on said measure of upstream congestion.

18. A method according to claim 1, further comprising a step of changing at least one of said first and second proportions in response to an indication from a receiver of a message that said information requires updating, said changing step comprising:

selecting a proportion of messages from a plurality of subsequently-sent messages, said proportion being dependent on a characteristic of an indication from said receiver;

assigning codepoints to said selected messages such as to cause said information to be updated.

19. A method according to claim 18 wherein said selected portion comprises the Explicit Congestion Notification field, and wherein said indication comprises a “

Congestion Experienced”

notification, and said step of selecting comprises selecting a message subsequent to receipt of said “

Congestion Experienced”

notification.

20. A method according to claim 16 wherein said selected portion comprises the Explicit Congestion Notification field, and wherein said indication comprises a plurality of echoes of a “

Congestion Experienced”

notification, and said step of selecting comprises determining an estimate for the number N of messages acknowledged between receipt of two successive echoes and selecting one message for approximately every N−

1 messages acknowledged.

21. A method according to claim 20, wherein said estimate is a moving average and said step of selecting consists of:

selecting one message for every echo received;

and selecting one extra message for every N−

1 echoes received.

22. A method according to claim 1, wherein one or both of said sets of codepoints consists of one codepoint.

23. A method according to claim 1 for performing the stateless extraction of separate sequences from repetitions of a single field whereby to communicate more than one signal at once.

24. A method according to claim 23, further comprising superposition of the results of each by various arithmetic combinations.

25. A method according to claim 1, at least one of said steps of assigning codepoints being performed by a node in said network when sending messages on a path traversing a plurality of nodes in the network.

26. A method according to claim 1, at least one of said steps of assigning codepoints being performed by a node in said network when forwarding messages on a path traversing a plurality of nodes in the network.

27. Apparatus for encoding information comprising means for carrying out the steps for implementing methods according to claim 1.