Systems and methods for improving network performance

• US 20050165948A1
• Filed: 01/08/2004
• Published: 07/28/2005
• Est. Priority Date: 01/08/2004
• Status: Active Grant

• Current Assignee: Avago Technologies General IP PTE Limited (Broadcom, Inc.)
• Original Assignee: LSI Corporation (Broadcom, Inc.)
• Inventors: Hatime, Hicham
• Granted Patent: US 7,310,682 B2
• Time in Patent Office: Days
• US Class Current: 709/235
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/11 Identifying congestion
• CPC Class Codes: H04L 47/12 Avoiding congestion; Recovering from congestion
• CPC Class Codes: H04L 47/193 at the transport layer, e.g. TCP related
• CPC Class Codes: H04L 47/25 with rate being modified by the source upon detecting a change of network conditions
• CPC Class Codes: H04L 47/283 in response to processing delays, e.g. caused by jitter or round trip time [RTT]
• CPC Class Codes: H04L 47/37 Slow start
• 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 69/163 In-band adaptation of TCP data exchange; In-band control procedures

Abstract

Systems and methods for improving network throughput in a packet ties to network communication environment. In one aspect here of, delays associated with retransmission of a packet or reduced to improve total performance and bandwidth utilization on the communication network medium. A roundtrip time parameter of TCP/IP communications networks may be adjusted according to actual measured response time of a successful packet transmission and used as a timeout value to detect a possible packet loss sooner than previous techniques. Another aspect hereof provides for dynamically adjusting estimated available bandwidth of the communication medium. The estimated available bandwidth is then useful to better avoid congestion on the medium and the resulting packet loss.

Claims

1. A method of reducing network congestion comprising the steps of:

transmitting multiple packets from a sending node to a receiving node over a network communication medium at a rate no faster than a predetermined maximum packet transmission rate value;

receiving multiple acknowledgements at the sending node from the receiving node;

measuring a transmission performance value based on transmission of a packet of said multiple packets and a corresponding acknowledgment of the multiple acknowledgements;

predicting an available bandwidth value based on at least two transmission performance values; and

adjusting the predetermined maximum packet transmission rate based on the predicted available bandwidth to reduce network congestion on the network communication medium.

2. The method of claim 1 wherein the step of predicting comprises:

determining a polynomial equation based upon the at least two transmission performance values; and

extrapolating from the polynomial equation to determine the predicted available bandwidth value.

3. The method of claim 2 wherein the step of determining comprises:

determining a cubic spline function based upon the at least two transmission performance values.

4. The method of claim 1 wherein the step of measuring comprises:

determining a time interval between a first transmitted packet and receipt of a corresponding acknowledgement message as the transmission performance value.

5. The method of claim 1 wherein the step of measuring comprises:

counting a number of additional packets transmitted between a first transmitted packet and receipt of a corresponding acknowledgement message as the transmission performance value.

6. A method for reducing re-transmission delays in a packetized network protocol comprising:

measuring a time between a successful packet transmission and receipt of a corresponding acknowledgement;

using the measured time as a timeout value for a subsequent packet transmission;

detecting expiration of the timeout value following the transmission of said the subsequent packet; and

retransmitting the subsequent packet in response to detecting expiration of the timeout value.

7. The method of claim 6 further comprising:

detecting receipt of two negative acknowledgement packets,

wherein the step of retransmitting is further responsive to the detection of receipt of two negative acknowledgement packets.

8. A method for reducing TCP/IP network congestion comprising:

estimating available bandwidth on the TCP/IP network based on a positively acknowledged packet; and

responsive to the estimated available bandwidth, dynamically adjusting at least one TCP/IP transmission parameter to reduce packet loss due to network congestion.

9. The method of claim 8 wherein the TCP/IP parameters may include one or more of slow start threshold, congestion window and roundtrip time.

10. The method of claim 9 wherein the step of estimating comprises:

determining a number of additional packets transmitted between a first transmitted packet and an acknowledgement message received corresponding to the first transmitted packet.

11. The method of claim 9 wherein the step of estimating comprises:

determining a time interval between a first transmitted packet and an acknowledgement message received corresponding to the first transmitted packet.

12. The method of claim 9 wherein the step of adjusting comprises adjusting the roundtrip time.

13. The method of claim 12 further comprising:

applying the adjusted roundtrip time to reduce retransmission delays on the TCP/IP network.

14. The method of claim 13 wherein the step of applying comprises:

detecting expiration of the roundtrip time following transmission of a packet without receiving an acknowledgement; and

retransmitting the packet in response to detection of the expired roundtrip time.

15. A system comprising:

a packet network communication medium;

a first network appliance communicatively coupled to the medium; and

one or more other network appliances communicatively coupled to the first network appliance via the medium,

wherein the first network appliance further comprises;

a predictive tuning element to dynamically predictively adapt protocol parameters of the first network appliance based on present and past measurements of throughput on the medium.

16. The system of claim 15 wherein the predictive tuning element further comprises:

a congestion avoidance element to dynamically adapt transmission protocol parameters to avoid congestion on the medium.

17. The system of claim 16 wherein the first network appliance is a TCP/IP protocol compatible network appliance and wherein the congestion avoidance element dynamically adapts the cwnd TCP/IP parameter to avoid congestion based on current and past measurement of throughput applied to a predictive mathematical model.

18. The system of claim 17 wherein the predictive mathematical model fits a cubic spline polynomial to the present and past throughput measurements to predict future throughput and corresponding adaptation of cwnd.

19. The system of claim 15 wherein the predictive tuning element further comprises:

a retransmission element to dynamically adapt retransmission protocol parameters to reduce latencies associated with detecting packet loss on the medium.

20. The system of claim 19 wherein the first network appliance is a TCP/IP protocol compatible network appliance and wherein the retransmissions element dynamically adapts the RTT TCP/IP parameter to reduce latencies based on current and past measurement of throughput applied to a predictive mathematical model.

21. The system of claim 20 wherein the predictive mathematical model fits a cubic spline polynomial to the present and past throughput measurements to predict future throughput and corresponding adaptation of RTT.

22. A computer readable storage medium tangibly embodying program instructions to perform a method of reducing network congestion, the method comprising the steps of:

transmitting multiple packets from a sending node to a receiving node over a network communication medium at a rate no faster than a predetermined maximum packet transmission rate value;

receiving multiple acknowledgements at the sending node from the receiving node;

measuring a transmission performance value based on transmission of a packet of said multiple packets and a corresponding acknowledgment of the multiple acknowledgements;

predicting an available bandwidth value based on at least two transmission performance values; and

adjusting the predetermined maximum packet transmission rate based on the predicted available bandwidth to reduce network congestion on the network communication medium.

23. The storage medium of claim 22 wherein the step of predicting comprises:

determining a polynomial equation based upon the at least two transmission performance values; and

extrapolating from the polynomial equation to determine the predicted available bandwidth value.

24. The storage medium of claim 23 wherein the step of determining comprises:

determining a cubic spline function based upon the at least two transmission performance values.

25. The storage medium of claim 22 wherein the step of measuring comprises:

determining a time interval between a first transmitted packet and receipt of a corresponding acknowledgement message as the transmission performance value.

26. The storage medium of claim 22 wherein the step of measuring comprises:

counting a number of additional packets transmitted between a first transmitted packet and receipt of a corresponding acknowledgement message as the transmission performance value.

27. A method for reducing re-transmission delays in a packetized network protocol comprising:

measuring a time between a successful packet transmission and receipt of a corresponding acknowledgement;

using the measured time as a timeout value for a subsequent packet transmission;

detecting expiration of the timeout value following the transmission of said the subsequent packet; and

retransmitting the subsequent packet in response to detecting expiration of the timeout value.

28. The method of claim 27 further comprising:

detecting receipt of two negative acknowledgement packets,

wherein the step of retransmitting is further responsive to the detection of receipt of two negative acknowledgement packets.