Congestion control for delay sensitive applications

• US 8,553,540 B2
• Filed: 04/16/2010
• Issued: 10/08/2013
• Est. Priority Date: 03/05/2010
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Mehrotra, Sanjeev, Wynn, Tong L., Li, Jin, Sengupta, Sudipta
• Primary Examiner(s): Pham, Chi
• Assistant Examiner(s): HOM, SHICK C
• Application Number:
  US12/762,016
• Publication Number:
  US 20110216648A1
• Time in Patent Office: 1,271 Days
• Field of Search: None
• US Class Current: 370/230
• CPC Class Codes: H04L 47/22 Traffic shaping
• CPC Class Codes: H04L 47/2416 Real-time traffic
• CPC Class Codes: H04L 47/25 with rate being modified by the source upon detecting a change of network conditions
• CPC Class Codes: H04L 47/29 using a combination of thresholds
• CPC Class Codes: H04L 47/30 in combination with information about buffer occupancy at either end or at transit nodes

Abstract

In various embodiments, methods and systems are disclosed for a hybrid rate plus window based congestion protocol that controls the rate of packet transmission into the network and provides low queuing delay, practically zero packet loss, fair allocation of network resources amongst multiple flows, and full link utilization. In one embodiment, a congestion window may be used to control the maximum number of outstanding bits, a transmission rate may be used to control the rate of packets entering the network (packet pacing), a queuing delay based rate update may be used to control queuing delay within tolerated bounds and minimize packet loss, and aggressive ramp-up/graceful back-off may be used to fully utilize the link capacity and additive-increase, multiplicative-decrease (AIMD) rate control may be used to provide fairness amongst multiple flows.

Claims

1. A method for transmitting data over a communications network, the method comprising:

queuing data packets for transmission over the communications network;

determining that a number of queued data packets is less than a predetermined queuing threshold and transmitting one of the data packets, wherein the predetermined queuing threshold indicates a maximum quantity of data that can be queued and not transmitted over the communications network; and

adjusting the transmission rate of the data packets based upon a congestion level of the communications network, the congestion level based at least in part on received acknowledgement of the transmitted data packets and determined as a function of a packet queuing delay, packet loss rate, and one way delay (OWD) trend estimated based on acknowledgement packets.

2. The method of claim 1, wherein the congestion level is determined as a function of a packet queuing delay.

3. The method of claim 2, wherein the packet queuing delay is computed by taking an average of the relative one way delay (ROWD) measurements of packets that have been acknowledged during a predetermined time interval.

4. The method of claim 1, wherein said adjusting further comprises transmitting the queued data packets and increasing the rate of transmission until congestion is detected.

5. The method of claim 4, further comprising:

when said congestion is detected, reducing said rate of transmission until said congestion is no longer detected.

6. The method of claim 1, wherein said packet queuing delay and said OWD trend are estimated based on a round-trip time (RTT) and a OWD, and said packet loss rate is estimated by:

computing a number of packets lost during a sliding window to compute an estimate of the current loss rate; and

smoothing the estimated current loss rate using an exponential weighted moving average of the current loss rate to obtain an estimate of the probability of loss rate, wherein packets are declared to be lost when no acknowledgement is received within a retransmission time-out (RTO) that is computed using RTT and the variance in RTT (RTTVAR).

7. The method of claim 3, wherein the predetermined time interval comprises a unit of time corresponding to the estimated round trip propagation time of the communications network.

8. The method of claim 3, wherein the ROWD is computed as ROWD=OWD−

OWD_min, wherein OWD is the actual one-way delay computed as a received time using a receiver'"'"'s clock minus the sent time using a sender'"'"'s clock, and OWD_min is the minimum observed OWD value.

9. The method of claim 1, further comprising:

classifying network utilization level into one of three zones;

Zone 1;

OWD is not increasing and the average packet queuing delay is less than a predetermined delay threshold,

Zone 2;

OWD is not increasing, packet loss is below a predetermined loss threshold, and the average queuing delay is greater than the predetermined delay threshold but less than an upper threshold, and

Zone 3;

OWD is increasing, the average queuing delay is greater than the upper threshold or packet loss above a packet loss threshold is accompanied by average queuing delay above a queuing delay threshold.

10. The method of claim 9, wherein the transmission rate (R_n) is updated using the current zone classification, the current average packet queuing delay estimate (δ

_avg), packet loss, and OWD trend (increasing or non-increasing), using

11. The method of claim 10, wherein the transmission rate is increased when the congestion level is in zone 1, the transmission rate is reduced when the congestion level is in zone 2 or zone 3, and the transmission rate is reduced at a greater rate in zone 3 than in zone 2.

12. The method of claim 2, wherein the packet queuing delay is computed using ROWD=OWD−

OWD_{min1} until T exceeds a predetermined threshold t3, whereupon ROWD=OWD−

OWD_{min2}) and OWD_{min1} is set to the previous value of ROWD, wherein OWD_{min1} is the minimum observed OWD from time “

t1”

to the current time T, and OWD_{min2} is the minimum observed OWD from time “

t2”

(where t2>

t 1 ) to T.

13. The method of claim 3, wherein the average of the relative one way delay is determined using a difference between an actual OWD and a minimum OWD measured over a predetermined number of OWD values, and the OWD is a difference between a packet received time using a receiver clock and a sent time using a sender clock.

17. The method of claim 1, further comprising computing a number of packets lost during a sliding window to compute an estimate of the current loss rate, wherein the sliding window has a size based on a transmission rate and an epoch length.

18. The method of claim 17, wherein the epoch length is based on an estimated round trip propagation time.

19. The method of claim 18, wherein, following each epoch, the epoch length is updated and the congestion level is classified into one of a plurality of zones.

14. A system configured to transmit data over a communications network, comprising:

at least one processor; and

at least one memory communicatively coupled to said at least one processor when the system is operational, the memory having stored therein computer-executable instructions that, upon execution by the processor, cause;

queuing data packets for transmission over the communications network;

determining that a number of queued data packets is less than a predetermined queuing threshold and transmitting one of the data packets, wherein the predetermined queuing threshold indicates a maximum quantity of data that can be queued and not transmitted over the communications network;

increasing a rate of transmission until congestion is detected, the congestion detected as a function of packet queuing delay, packet loss rate, and one-way delay (OWD) trend and based at least in part on received acknowledgement of the transmitted data packets; and

reducing said rate of transmission until said congestion is no longer detected;

wherein the transmission rate (R_n) is updated using the OWD trend, packet loss, the current average packet queuing delay estimate (δ

_avg), and the current zone classification which is based on the current average packet queuing delay estimate (δ

_avg), packet loss, and OWD trend, using

15. The system of claim 14, further comprising classifying network utilization level into one of three zones:

(1) OWD is not increasing and the average packet queuing delay is less than a predetermined delay threshold,

(2) OWD is not increasing, packet loss is below a predetermined loss threshold, and the average queuing delay is greater than the predetermined delay threshold but less than an upper threshold, and

(3) OWD is increasing, the average queuing delay is greater than the upper threshold or packet loss above a packet loss threshold is accompanied by average queuing delay above a queuing delay threshold, wherein the transmission rate is increased when the congestion level is in zone 1, the transmission rate is reduced when the congestion level is in zone 2 or zone 3, and the transmission rate is reduced at a greater rate in zone 3 than in zone 2.

16. A computer readable storage device storing thereon computer executable instructions, said instructions comprising:

instructions for queuing data packets for transmission over the communications network;

instructions for determining that a number of queued data packets is less than a predetermined queuing threshold and transmitting one of the data packets, wherein the predetermined queuing threshold indicates a maximum quantity of data that can be queued and not transmitted over the communications network;

instructions for increasing a rate of transmission until congestion is detected, the congestion detected as a function of packet queuing delay, packet loss rate, and one-way delay (OWD) trend;

instructions for reducing said rate of transmission until said congestion is no longer detected; and

instructions for classifying network utilization level into one of three zones;

(1) OWD is not increasing and the average packet queuing delay is less than a predetermined delay threshold,

(2) OWD is not increasing, packet loss is below a predetermined loss threshold, and the average queuing delay is greater than the predetermined delay threshold but less than an upper threshold, and

(3) OWD is increasing, the average queuing delay is greater than the upper threshold or packet loss above a packet loss threshold is accompanied by average queuing delay above a queuing delay threshold, wherein the transmission rate is increased when the congestion level is in zone 1, the transmission rate is reduced when the congestion level is in zone 2 or zone 3, and the transmission rate is reduced at a greater rate in zone 3 than in zone 2;

wherein;

the congestion is determined as a function of a packet queuing delay, packet loss rate, and OWD trend, and

the transmission rate (R_n) is updated using the current zone classification, the current average packet queuing delay estimate (δ

_avg), packet loss, and OWD trend (increasing or non-increasing), using