METHOD AND APPARATUS FOR CONGESTION CONTROL

• US 20120155258A1
• Filed: 06/10/2010
• Published: 06/21/2012
• Est. Priority Date: 06/18/2009
• Status: Active Grant

• Current Assignee: RPX Corporation
• Original Assignee: Alcatel-Lucent SA (Nokia Corporation)
• Inventors: De Vleeschauwer, Danny, Laevens, Koenraad, Baccelli, Francois, Hong, Dohy
• Granted Patent: US 8,867,347 B2
• Time in Patent Office: Days
• US Class Current: 370/230
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/127 by using congestion prediction
• CPC Class Codes: H04L 47/19 at layers above the network layer network arrangements for networked applications for scheduling or organising the servicing of application requests H04L67/60
• CPC Class Codes: H04L 47/2416 Real-time traffic
• CPC Class Codes: H04L 47/38 by adapting coding or compression rate
• CPC Class Codes: H04L 47/70 Admission control; Resource allocation
• CPC Class Codes: H04L 47/724 at intermediate nodes, e.g. resource reservation protocol [RSVP]
• CPC Class Codes: H04L 47/781 Centralised allocation of resources
• CPC Class Codes: H04L 47/801 Real time traffic
• CPC Class Codes: H04L 47/822 Collecting or measuring resource availability data
• CPC Class Codes: H04L 47/83 based on usage prediction

Abstract

A method for control of congestion of traffic, being categorized according to a number of layers (L), in a network node, comprises a step of monitoring the amount (n(t)) of incoming flows to this node such that, based upon this amount (n(t)) of incoming flows and on a current number of accepted layers (l(t)), a next number of accepted layers (l(t+1)) for entry into said node is determined by consultation of an action table (T). In an enhanced embodiment the action table is adjusted based on the observed traffic during an observation period.

Claims

1. Method for control of congestion of traffic in a network node, said traffic being categorized according to a number of layers (L), said method comprising a step of monitoring the amount (n(t)) of incoming flows to this node such that, based upon this amount (n(t)) of incoming flows and on a current number of accepted layers (l(t)), a next number of accepted layers (l(t+1)) for entry into said node is determined by consultation of an action table (T).

2. Method according to claim 1 wherein said action table is dynamically adjusted based on observed traffic towards said node.

3. Method according to claim 2 wherein said action table is dynamically adjusted at regular intervals based upon traffic observed during an observation period, and whereby a Markov Decision Process is used for determining an optimum action table.

4. Method according to claim 1 wherein the action table is locally stored within said node.

5. Method according to claim 1 wherein said action table is centrally calculated by a network congestion controller and further communicated to said node.

6. Method according to claim 1 further comprising a step of communicating, between different nodes implementing said method, the resulting action as determined by consultation of said action table, whereby, in case conflicting actions arise, a further heuristic control for adapting the next number of accepted layers, per node, is executed within said nodes.

7. Method according to claim 1 wherein said action table is centrally updated by a network congestion controller.

8. Congestion control device (CCIM;

CCIM′

;

CCAN1;

CCAN2;

CCAN3) for control of congestion of traffic categorized according to a number of layers (L) entering a network node (IM;

AN1;

AN2;

AN3), said congestion control device (CCIM;

CCIM′

) being adapted to monitor the amount (n(t)) of incoming flows to this node and to determine, based upon this amount (n(t)) of incoming flows, based upon an action table (T) and on a current number of accepted layers (l(t)), a next number of accepted layers (l(t+1)), such as to only allow packets identified by a layer number related to said number of accepted layers to enter said node.

9. Congestion control device (CCIM′

) according to claim 8 further being adapted to dynamically adjust said action table based on observed traffic towards said node during an observation period.

10. Congestion control device (CCIM′

) according to claim 9 being adapted to dynamically adjust said action table at regular intervals based upon traffic observed during said observation period, by using a Markov Decision for thereby determining an optimum action table.

11. Congestion control device (CCIM) according to claim 9 further being adapted to communicate with another congestion control device in another node, the resulting action as determined by consultation of said action table, and further being adapted to, in case conflicting actions arise, implement a further heuristic control for adapting the next number of accepted layers.

12. Network node (AN1;

AN2;

AN3;

IM) wherein said network node comprised a congestion control device (CCIM;

CCIM′

;

CCAN2;

CCAN3;

CCIM) according to claim 8.

13. Network congestion control device (NC) for communicating with a congestion control device (CCIM) according to claim 8, said network congestion control device (NC) being adapted to determine said action table (T) and to communicate said action table to said congestion control device (CCIM).

14. Network congestion control device (NC) according to claim 13, further being adapted to update said action table based on observed traffic, and to provide said updated action table to said congestion control device (CCIM).