Methods and apparatus for use in packet-switched data communication networks

• US 20060045011A1
• Filed: 11/26/2003
• Published: 03/02/2006
• Est. Priority Date: 11/26/2002
• Status: Abandoned Application

• Current Assignee: King's College London
• Original Assignee: King's College London
• Inventors: Aghvami, Abdol Hamid, Friderikos, Vasilis Athanasios
• Application Number:
  US10/536,380
• Publication Number:
  US 20060045011A1
• Time in Patent Office: Days
• US Class Current: 370/230
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/193 at the transport layer, e.g. TCP related
• CPC Class Codes: H04L 47/26 using explicit feedback to the source, e.g. choke packets
• 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 47/323 Discarding or blocking control packets, e.g. ACK packets

Abstract

A method of reducing packet congestion at a network node in a packet-switched data communication network, which method comprises the steps of marking one or more packets in said queue, wherein a probability of marking a packet in the queue is higher for a first proportion of the packets than for a second proportion of the packets, one or more packets of said first proportion having spent less time on said network than one or more packets of said second proportion.

Claims

1. A method of reducing packet congestion at a network node in a packet-switched data communication network, which method comprises the steps of marking one or more packets in a queue, wherein a probability of marking a packet in the queue is higher for a first proportion of the packets than for a second proportion of the packets, one or more packets of said first proportion having spent less time on said network than one or more packets of said second proportion.

2. A method as claimed in claim 1, wherein said time spent on the network is indicated by the number of network nodes crossed by each packet, the method further comprising the step of determining said probability based on said number of network nodes.

3. A method as claimed in claim 2, wherein each packet comprises an Internet Protocol (IP) header, the method further comprising the step of obtaining the number of network nodes by reading the Time To Live (TTL) field in an IPv4 header, or from the Hop Limit field (HL) in an IPv6 header of each packet.

4. A method as claimed in claim 2, further comprising the steps of examining the queue to determine a maximum number of network nodes and a minimum number of network nodes crossed by packets therein, and determining a probability of being marked for each network node number between said maximum and said minimum.

5. A method as claimed in claim 1, wherein said probability varies as a function of the time each packet has spent crossing the network or as a function of the number of network nodes.

6. A method as claimed in claim 5, wherein said function is of substantially linear form and the probability is inversely proportional to the time each packet has spent crossing the network or the number of network nodes crossed by each packet.

7. A method as claimed in claim 1, wherein said method comprises the steps of:

(a) receiving a packet on said network node;

(b) adding said packet to said queue;

(c) determining whether or not the addition of said packet to said queue causes a congestion condition in said network node;

(d) if there is a congestion condition, determining a hop number h of said packet corresponding to the number of network nodes crossed by said packet prior to reaching said network node;

(e) storing said packet in memory;

(f) examining each packet in said queue to determine the maximum hop number, h_max, and the minimum hop number, h_min, present therein;

(g) determining a minimum marking probability (λ

_hmin) for packets having said minimum hop number, obtainable from the equation;

${\mathrm{\lambda <br><br>}}_{h_{\min }}^{-1}=1+\left(N-1\right)<br><br>{\left(\frac{{\mathrm{ah}}_{\min }+b}{{\mathrm{ah}}_{\max }+b}\right)}^2+\left(N-2\right)<br><br>\frac{h_{\max }}{h_{\max }-h_{\min }}<br><br>\left(1-{\left(\frac{{\mathrm{ah}}_{\min }+b}{{\mathrm{ah}}_{\max }+b}\right)}^2\right)-\frac{1}{h_{\max }-h_{\min }}<br><br>\left(1-{\left(\frac{{\mathrm{ah}}_{\min }+b}{{\mathrm{ah}}_{\max }+b}\right)}^2\right)<br><br>\underset{i=1}{\overset{N-2}{\sum <br><br>}}<br><br>\text{\hspace{1em}<br><br>}<br><br>h_i,$

 

determining a maximum marking probability (λ

_hmax) for packets having said maximum hop number, obtainable from the equation;

${\mathrm{\lambda <br><br>}}_{h_{\max }}={{\mathrm{\lambda <br><br>}}_{h_{\min }}<br><br>\left(\frac{{\mathrm{ah}}_{\min }+b}{{\mathrm{ah}}_{\max }+b}\right)}^2$

 

determining an individual marking probability (λ

_hi) for each packet having hop number i in said packet queue, obtainable from the equation;

${\mathrm{\lambda <br><br>}}_{h_i}=-\frac{{\mathrm{\lambda <br><br>}}_{h_{\min }}-{\mathrm{\lambda <br><br>}}_{h_{\max }}}{h_{\max }-h_{\min }}<br><br>h_i+{\mathrm{\lambda <br><br>}}_{h_{\max }}+h_{\max }<br><br>\frac{{\mathrm{\lambda <br><br>}}_{h_{\min }}-{\mathrm{\lambda <br><br>}}_{h_{\max }}}{h_{\max }-h_{\min }}$

 

where N is the number of different hop numbers in the queue (i.e. h_max−

h_min) and a and b are coefficients depending on the relationship between round trip time and propagation delay of packets passing through said network node, obtainable from the equation τ

_i=ah_i+b, where τ

_i is the round trip time for packets with hop number i;

(h) normalising λ

_hmax, λ

_hmin and λ

_hi to form respective normalised marking probabilities for each hop number as follows;

if there are n_hmin, n₁, . . . n_i . . . n<sub>N−

2</sub>, n_hmax packets of each hop number in said packet queue and if the percentage of packets per hop number is given by ν

_hmin, ν

₁, . . . ν

_i . . . ν

<sub>N−

2</sub>, ν

_hmax, where ν

_i=n_i/n and n=n_hmin+n₁+ . . . +n_i+ . . . +n<sub>N−

2</sub>+n_hmax, then said normalised marking probabilities are obtainable from;

${\mathrm{\pi <br><br>}}_{h_{\min }}={\mathrm{\lambda <br><br>}}_{h_{\min }}·<br><br>\frac{{\mathrm{??}}_{h_{\min }}}{\mathrm{\pi <br><br>}},{\mathrm{\pi <br><br>}}_1={\mathrm{\lambda <br><br>}}_1<br><br>\frac{{\mathrm{??}}_1}{\mathrm{\pi <br><br>}},\mathrm{\dots <br><br>}<br><br>\text{\hspace{1em}<br><br>},{\mathrm{\pi <br><br>}}_{h_{\max }}={\mathrm{\lambda <br><br>}}_{h_{\max }}·<br><br>\frac{{\mathrm{??}}_{h_{\max }}}{\mathrm{\pi <br><br>}}$

 

where π

=λ

_hminν

_hmin+λ

₁ν

₁+ . . . +λ

<sub>N−

2</sub>ν

<sub>N−

2</sub>+λ

_hmaxν

_hmax; and

(i) marking packets of each hop number in said queue according to the corresponding normalised marking probability.

8. A method as claimed in claim 1, wherein packets in said first proportion have a substantially constant first probability of being marked and packets in said second proportion have a substantially constant second probability of being marked lower than said first probability.

9. A method as claimed in claim 8, wherein the packets in the queue are divided into said first and second proportions by a threshold based upon the mean number of networks nodes crossed by the packets in the queue.

10. A method as claimed in claim 9, wherein the threshold is approximately equal to the mean number of hops in the queue plus one standard deviation.

11. A method as claimed in claim 8, wherein said method comprises the steps of:

(a) receiving a packet on said network node;

(b) adding said packet to said queue;

(c) determining whether or not the addition of said packet to said queue causes a congestion condition in said network node;

(d) if there is a congestion condition, determining a hop number h of said packet corresponding to the number of network nodes crossed by said packet prior to reaching said network node;

(e) storing said packet in memory;

(f) examining each packet in said queue to determine the maximum hop number, h_max, and the minimum hop number, h_min, present therein;

(g) determining a hop number threshold θ

from the distribution of hop numbers in said queue, where θ

is one standard deviation;

(h) determining a first marking probability λ

<sub>θ

−</sub>

for packets having a hop number less than said hop number threshold θ

, obtainable from the equation;

${\mathrm{\lambda <br><br>}}_{{\mathrm{\theta <br><br>}}_{-}}={\left[1+{\left(\frac{{\mathrm{ah}}_{\min }+b}{{\mathrm{ah}}_{\max }+b}\right)}^2\right]}^{-1},$

 

determining a second marking probability λ

<sub>θ

+</sub> for packets having a hop number greater than said hop number threshold θ

, obtainable from the equation;

λ

<sub>θ

+</sub>=1−

λ

<sub>θ

−</sub>

,

where a and b are coefficients depending on the relationship between round trip time and propagation delay of packets passing through said network node, obtainable from the equation τ

_i=ah_i+b, where τ

_i is the round trip time for packets with hop number i;

(i) normalising said first and second marking probabilities to form respective first and second normalised marking probabilities, π

<sub>θ

−</sub>

and π

<sub>θ

+</sub>, obtainable from;

${\mathrm{\pi <br><br>}}_{{\mathrm{\theta <br><br>}}_{-}}={\left[1+{\left(\frac{{\mathrm{ah}}_{\min }+b}{{\mathrm{ah}}_{\max }+b}\right)}^2\right]}^{-1}<br><br>\frac{{\mathrm{??}}_{{\mathrm{\theta <br><br>}}_{-}}}{{\mathrm{\lambda <br><br>}}_{h_{\min }}<br><br>{\mathrm{??}}_{{\mathrm{\theta <br><br>}}_{-}}+{\mathrm{\lambda <br><br>}}_{h_{\max }}<br><br>{\mathrm{??}}_{{\mathrm{\theta <br><br>}}_{+}}}<br><br>\mathrm{and}$${\mathrm{\pi <br><br>}}_{{\mathrm{\theta <br><br>}}_{+}}={{\left(\frac{{\mathrm{ah}}_{\min }+b}{{\mathrm{ah}}_{\max }+b}\right)}^2<br><br>\left[1+{\left(\frac{{\mathrm{ah}}_{\min }+b}{{\mathrm{ah}}_{\max }+b}\right)}^2\right]}^{-1}<br><br>\frac{{\mathrm{??}}_{{\mathrm{\theta <br><br>}}_{+}}}{{\mathrm{\lambda <br><br>}}_{h_{\min }}<br><br>{\mathrm{??}}_{{\mathrm{\theta <br><br>}}_{-}}+{\mathrm{\lambda <br><br>}}_{h_{\max }}<br><br>{\mathrm{??}}_{{\mathrm{\theta <br><br>}}_{+}}};<br><br>\mathrm{and}$

(j) marking packets in said queue according to said first and second marking probabilities π

<sub>θ

−</sub>

and π

<sub>θ

+</sub>.

12. A method as claimed in claim 1, further comprising the step of initiating the method in response to an indication of congestion at said network node caused by a queue of packets awaiting processing at said network node.

13. A method as claimed in claim 12, wherein said indication is provided by a method employing Random Early Detection (RED).

14. A method as claimed in claim 1, wherein the step of marking a packet comprises dropping the packet, setting the Explicit Congestion Notification bit in the IP header or performing any other step that identifies congestion to a transport protocol used by the intended recipient of the packet.

15. A method as claimed in claim 12, further comprising the step of repeating the method upon receipt of a further indication of congestion at the network node.

16. A computer program product storing computer executable instructions in accordance with a method of claim 1.

17. A computer program product as claimed in claim 16, embodied on a record medium, in a computer memory, in a read-only memory, or on an electrical carrier signal.

18. A network node for use in a packet-switched data communication network, which network node comprises an interface for receiving packets from other network nodes, a routing table for determining the identity of a subsequent network node to which each packet should be sent, a memory for temporary storage of packets and a processor for forwarding each packet to the subsequent network node, wherein said memory stores computer executable instructions in accordance with a method as claimed in claim 1 for execution by said processor.

19. A network node as claimed in claim 18, embodied in an OSI layer 4 routing device, for example a router and a gateway router.

20. A network node as claimed in claim 18, embodied in a wireless communication device, for example a hand-held wireless device.

21. A packet-switched data communication network comprising a plurality of network nodes, each of which can send and receive packets of data to and from other network nodes, wherein one or more network nodes is in accordance with claim 1.

23. A method of reducing the aggregate battery power consumption by network nodes in an ad-hoc computer network, which method comprises the steps of:

(1) using at least one of the network nodes to route data between communicating network nodes of the ad-hoc computer network; and

(2) at one or more routing network node using a method in accordance with claim 1 to mark packets passing therethrough, whereby packets in flows of data with a roundtrip time that is high compared to other flows of data through that routing network node have a lower probability of re-transmission across the ad-hoc network, thereby reducing the aggregate battery power consumption of routing network nodes in the ad-hoc computer network.

22. At a network node in a packet-switched data communication network, a method of initiating a reduction in the transmission rate of packets from a first host transmitting data over that network, which method comprises the steps of:

(1) receiving a packet directly or indirectly from said first host destined for a second host reachable directly or indirectly from said network node; and

(2) either marking or not marking said packet, a probability of marking the packet being determined on the basis of the time said packet has spent reaching said network node over at least a part of said network;

wherein marking of said packet serves to cause a subsequent reduction of said transmission rate from said first host.