Reliable transport protocol providing receiver-based congestion control

• US 7,995,475 B2
• Filed: 10/31/2008
• Issued: 08/09/2011
• Est. Priority Date: 10/31/2007
• Status: Active Grant

• Current Assignee: Architecture Technology Corporation
• Original Assignee: Architecture Technology Corporation
• Inventors: Ramanujan, Ranga S., Carl, Richard H., Trent, Barry, Mandy, Edward, Sanders, Clint, Bonney, Jordan C.
• Primary Examiner(s): Pham, Chi H
• Assistant Examiner(s): Boakye, Alexander
• Application Number:
  US12/262,624
• Publication Number:
  US 20100039937A1
• Time in Patent Office: 1,012 Days
• Field of Search: 370/232, 370/229, 370/233, 370/234, 370/235, 370/237, 370/236, 370/253, 370/282, 370/392, 370/394, 370/395.52, 370/466, 370/467, 709/223, 714/752
• US Class Current: 370/232
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/12 Avoiding congestion; Recovering from congestion
• CPC Class Codes: H04L 47/263 Rate modification at the source after receiving feedback
• CPC Class Codes: Y02D 30/50 in wire-line communication networks, e.g. low power modes or reduced link rate

Abstract

The present application relates to an implementation of a reliable transport protocol that provides receiver-based congestion control within a computing system. An exemplary system includes a first network device and a second network device within a network. During a handshake procedure, a transmission rate at which data can be sent as well as an acceptable corruptive loss rate for a data path between the first network device and the second network device in the network is determined. The second network device then receives data sent from the first network device at the transmission rate. When the acceptable corruptive loss rate for received data has been exceeded, the second network device determines an amount by which to reduce the transmission rate, and then sends a notification to the first network device indicating the amount by which to reduce the transmission rate for subsequently transmitted data.

Claims

1. A method comprising:

determining, during a handshake procedure between a first network device and a second network device in a network, a transmission rate at which data can be sent as well as an acceptable corruptive loss rate for a data path between the first network device and the second network device;

receiving, by the first network device, data sent from the second network device at the transmission rate across the data path;

determining an actual loss rate for the received data across the data path; and

when the actual loss rate for the received data exceeds the acceptable corruptive loss rate, determining an amount by which to reduce the transmission rate and sending a notification from the first network device to the second network device indicating the amount by which to reduce the transmission rate for subsequently transmitted data.

2. The method of claim 1, wherein determining an amount by which to reduce the transmission rate comprises subtracting the acceptable corruptive loss rate from the actual loss rate.

3. The method of claim 1, wherein sending a notification comprises sending a notification indicating the amount by which to reduce the transmission rate, wherein the amount is a percentage that is equivalent to a difference between the acceptable corruptive loss rate and the actual loss rate.

4. The method of claim 1, wherein determining an actual loss rate comprises comparing a number of packets actually received by the first network device with a number of packets alleged to have been sent by the second network device over a defined period of time.

5. The method of claim 1, further comprising:

transmitting data from the first network device to the second network device at the transmission rate;

receiving, by the first network device, a notification from the second network device indicating an amount by which to reduce the transmission rate;

reducing the transmission rate of the first network device in accordance with the notification; and

transmitting data to the second network device at the reduced transmission rate.

6. The method of claim 5, further comprising:

after transmitting data from the first network device to the second network device at the reduced transmission rate, increasing the transmission rate a defined amount when the first network device does not subsequently receive a second notification from the second network device indicating a second amount by which to reduce the transmission rate after a period of time; and

transmitting data from the first network device to the second network device at the increased transmission rate.

7. The method of claim 6, further comprising:

repeating the increasing of the transmission rate the defined amount after the period of time has elapsed again, until the increased transmission rate equals the transmission rate determined during the handshake procedure.

8. The method of claim 1, wherein determining, during a handshake procedure, an acceptable corruptive loss rate for a data path comprises:

sending, from the first network device to the second network device, a first handshake packet to request a data connection;

receiving, by the first network device, a second handshake packet from the second network device, the second handshake packet containing a network descriptor associated with the second network device;

identifying an acceptable corruptive loss rate for data transmission associated with the second network device using the received network descriptor associated with the second network device;

selecting the acceptable corruptive loss rate for data transmission across the path in accordance with the acceptable corruptive loss rate associated with the second network device and an acceptable corruptive loss rate associated with the first network device;

defining a path descriptor that includes the selected acceptable corruptive loss rate to be used by both the first network device and the second network device for data transmissions across the path; and

sending a third handshake packet from the first network device to the second network device, the third handshake packet containing the defined path descriptor.

9. The method of claim 8, wherein selecting the acceptable corruptive loss rate comprises:

setting the acceptable corruptive loss rate for data transmissions across the path as the higher of the acceptable corruptive loss rate associated with the second network device and the acceptable corruptive loss rate associated with the first network device.

10. The method of claim 8, wherein the network descriptor further defines a maximum transmission rate for data transmission associated with the second network device, and wherein the method further comprises:

identifying a maximum transmission rate associated with the first network device; and

wherein defining a path descriptor further comprises defining the path descriptor to include a transmission rate that is the lower of the maximum transmission rate associated with the first network device and the maximum transmission rate associated with the second network device.

11. The method of claim 8, wherein defining a path descriptor comprises:

defining a plurality of path descriptors, each of the plurality of path descriptors defining a path for information flow between the first network device and the second network device and an acceptable corruptive loss rate associated with the path.

12. The method of claim 11, further comprising:

determining that the plurality of different paths for information flow between the first network device and the second network device includes a number of paths equivalent to the higher of a number of ports on the first network device and a number of ports on the second network device.

13. The method of claim 11, further comprising:

receiving, by the first network device, data transmitted from the second network device across each of the plurality of different paths as separate microflows of data; and

assembling data from each of the microflows into a macroflow by using sequence numbers associated with the data to put the data in order.

14. The method of claim 1, wherein determining, during a handshake procedure between a first network device and a second network device, an acceptable corruptive loss rate further comprises determining the acceptable corruptive loss rate in terms of a number of packets lost over a period of time.

15. The method of claim 1, wherein determining, during a handshake procedure between a first network device and a second network device, an acceptable corruptive loss rate further comprises determining the acceptable corruptive loss rate in terms of a number of bits lost over a period of time.

16. A method comprising:

determining, during a handshake procedure, a transmission rate at which data can be sent as well as an acceptable corruptive loss rate for a data path between a first network device and a second network device in a network;

transmitting data from the second network device to the first network device across the data path at the transmission rate;

receiving, with the second network device from the first network device, a notification indicating that the acceptable corruptive loss rate has been exceeded, the notification further indicating an amount by which to reduce the transmission rate; and

reducing the transmission rate according to the amount indicated in the notification for subsequently transmitted data.

17. The method of claim 16, further comprising:

transmitting data from the second network device to the first network device at the reduced transmission rate.

18. The method of claim 17, further comprising:

after transmitting data from the second network device to the first network device at the reduced transmission rate, increasing the transmission rate a defined amount when the second network device does not subsequently receive a second notification from the first network device indicating a second amount by which to reduce the transmission rate after a period of time;

transmitting data from the second network device to the first network device at the increased transmission rate; and

repeating the increasing of the transmission rate the defined amount after the period of time has elapsed again, until the increased transmission rate equals the transmission rate determined during the handshake procedure.

19. The method of claim 16, wherein transmitting data from the second network device to the first network device includes transmitting information identifying a number of packets of data sent from the second network device to the first network device.

20. The method of claim 19, wherein transmitting information identifying a number of packets comprises transmitting a sequence number with each of the packets.

21. The method of claim 19, wherein transmitting data comprises transmitting packets, wherein the notification indicates that an actual loss rate exceeds the acceptable corruptive loss rate for packet data received by the first network device, and wherein the actual loss rate is based on a comparison of a number of data packets received by the first network device with the number of data packets identified to have been sent from the second network device to the first network device over a defined period of time.

22. The method of claim 16, wherein determining, during a handshake procedure, a transmission rate and an acceptable corruptive loss rate for a data path comprises:

receiving, from the first network device by the second network device, a first handshake packet that requests a data connection;

obtaining a network descriptor associated with the second network device;

sending, to the first network device, a second handshake packet from the second network device, the second handshake packet containing the network descriptor associated with the second network device;

receiving a third handshake packet from the first network device with the second network device, the third handshake packet containing a path descriptor that defines an acceptable corruptive loss rate and a transmission rate;

setting the acceptable corruptive loss rate for the data path as the acceptable corruptive loss rate of the path descriptor; and

setting the transmission rate for the data path as the transmission rate of the path descriptor.

23. A non-transitory computer-readable medium comprising instructions for causing a programmable processor to:

determine, during a handshake procedure between a first network device and a second network device in a network, a transmission rate at which data can be sent as well as an acceptable corruptive loss rate for a data path between the first network device and the second network device;

receive, by the first network device, data sent from the second network device at the transmission rate across the data path;

determine an actual loss rate for the received data across the data path; and

when the actual loss rate for the received data exceeds the acceptable corruptive loss rate, determine an amount by which to reduce the transmission rate and send a notification from the first network device to the second network device indicating the amount by which to reduce the transmission rate for subsequently transmitted data.

24. A non-transitory computer-readable medium comprising instructions for causing a programmable processor to:

determine, during a handshake procedure, a transmission rate at which data can be sent as well as an acceptable corruptive loss rate for a data path between a first network device and a second network device in a network;

transmit data from the second network device to the first network device across the data path at the transmission rate;

receive, with the second network device from the first network device, a notification indicating that the acceptable corruptive loss rate has been exceeded, the notification further indicating an amount by which to reduce the transmission rate; and

reduce the transmission rate according to the amount indicated in the notification for subsequently transmitted data.

25. A network device comprising:

a network interface;

a processor; and

a module executable by the processor, the module comprising;

a handshake module executable by the processor to determine, during a handshake procedure between the network device and a second network device, a transmission rate at which data can be sent as well as an acceptable corruptive loss rate for a data path between the network device and the second network device; and

a transceiver module executable by the processor to receive data sent from the second network device at the transmission rate across the data path using the network interface, and to determine an actual loss rate for the received data across the data path by comparing a number of packets actually received with a number of packets alleged to have been sent by the second network device over a defined period of time,

wherein when the actual loss rate for the received data exceeds the acceptable corruptive loss rate, the transceiver module sends a notification to the second network device indicating an amount by which to reduce the transmission rate for subsequently transmitted data.

26. The network device of claim 25, further comprising:

a computer-readable medium to store the module as a set of executable instructions.

27. The network device of claim 25, further comprising a plurality of network interfaces, wherein:

the handshake module further determines a plurality of paths between the network device and the second network device, an acceptable corruptive loss rate for each of the plurality of paths, and a maximum transmission rate for each of the plurality of paths; and

the transceiver module receives data over each of the plurality of paths as a plurality of microflows, and reassembles microflow data received from the paths to form a macroflow in accordance with sequence numbers of packets of data received from the microflows.

28. The network device of claim 25, wherein the handshake module determines the acceptable corruptive loss rate in terms of a number of packets lost over a period of time.

29. The network device of claim 25, wherein the handshake module determines the acceptable corruptive loss rate in terms of a number of bits lost over a period of time.

30. A network device comprising:

a network interface;

a processor;

a handshake module executable by the processor to determine, during a handshake procedure between the network device and a second network device, a transmission rate at which data can be sent as well as an acceptable corruptive loss rate for a data path between the network device and the second network device; and

a transceiver module executable by the processor to transmit data to the second network device across the data path at the transmission rate using the network interface, and to receive from the second network device a notification indicating that the acceptable corruptive loss rate has been exceeded, the notification further indicating an amount by which to reduce the transmission rate, and

wherein the transceiver module reduces the transmission rate according to the amount indicated in the notification for subsequently transmitted data.