Method for accelerating TCP flows in communication networks with dynamic control
View Patent ↗Predictions of congestion conditions for a traffic stream in a communication network are applied to modify an initial congestion window size for the traffic stream; and dynamic bandwidth control is thereafter applied to the traffic stream. This dynamic bandwidth control may include modulating inter-packet bandwidths of the traffic stream according to a capacity of a bottleneck in a communication path through which the traffic stream passes in the communication network. The predictions of congestion conditions may be based on monitoring packet losses and/or round trip times within the communication network for a selected period of time. The monitoring may be performed on at least one of a traffic stream-by traffic stream basis, a connection-by-connection basis, a link-by-link basis, or a destination-by-destination basis.
1. A method, comprising:
applying predictions of congestion conditions for a traffic stream in a communication network to increase an initial congestion window size for the traffic stream up to an advertised window size of a client receiving the traffic stream; and
applying dynamic bandwidth control to the traffic stream by modulating bandwidths of the traffic stream according to a capacity of a bottleneck in a communication path through which the traffic stream passes in the communication network, wherein the dynamic bandwidth control is concurrently applied to the traffic stream to ensure that congestion is not induced by applying predictions of congestion conditions to the traffic stream.
2. The method of claim 1 wherein the predictions of congestion conditions are based on one of: (1) monitoring packet losses within the communication network, or (2) monitoring packet round trip time in the communication network.
3. The method of claim 2 wherein the monitoring is performed on at least one of a traffic stream-by traffic stream basis, a connection-by-connection basis, a link-by-link basis, or a destination-by-destination basis.
4. The method of claim 3 wherein the monitoring is performed for a period between 0 and 100 seconds.
5. The method of claim 4 wherein the monitoring is performed for a period of time between 30 and 100 seconds.
6. The method of claim 4 wherein the monitoring is performed for a period of time between 50 and 100 seconds.
7. The method of claim 4 wherein the monitoring is performed for a period of time between 60 and 100 seconds.
8. A method comprising:
setting an initial congestion window for a traffic stream in a communication network according to predicted congestion conditions for that traffic stream, increased up to an advertised window size of a client receiving the traffic stream; and
rate limiting the traffic stream to an effective bandwidth associated with a potentially congested bottleneck in a communication path over which the traffic stream is transmitted, wherein rate limiting the traffic stream acts as a back up to ensure that congestion is avoided between periods of setting the congestion window size.
9. The method of claim 8 wherein the rate limiting comprises setting a minimum time spacing between packets within the traffic stream.
10. The method of claim 8 wherein the rate limiting comprises setting the effective bandwidth equal to a maximum transfer rate allowed by the potentially congested bottleneck in the communication path.
11. The method of claim 8 wherein the rate limiting is applied using a feedback control process to modulate bandwidths in the traffic stream.
12. The method of claim 11 wherein the feedback control process is applied at a control node upstream of the potentially congested bottleneck in the communication path.
13. The method of claim 8 wherein the predicted congestion conditions are based on one of: (1) monitoring packet losses within the communication network, or (2) monitoring packet round trip time items in the communication network.
14. The method of claim 13 wherein the monitoring is performed for a period between 0 and 100 seconds.
15. The method of claim 14 wherein the monitoring is performed for a period of time between 30 and 100 seconds.
16. The method of claim 14 wherein the monitoring is performed for a period of time between 50 and 100 seconds.
17. The method of claim 14 wherein the monitoring is performed for a period of time between 60 and 100 seconds.
18. A communication network comprising one or more communication paths between one or more content sources and one or more clients, at least one of the communication paths including a control node configured to set an initial congestion window for a traffic stream transmitted over the at least one communication path according to predicted congestion conditions for that traffic stream, increased up to an advertised window size of a client receiving the traffic stream and to rate limit the traffic stream to an effective bandwidth associated with a potentially congested bottleneck in the at least one communication path over which the traffic stream is transmitted, wherein rate limiting the traffic stream acts as a back up to ensure that congestion is avoided between periods of setting the congestion window size.
19. The network of claim 18 wherein the control node is configured to rate limit the traffic stream by setting a minimum time spacing between packets within the traffic stream.
20. The network of claim 18 wherein the control node is configured to rate limit the traffic stream by setting the effective bandwidth equal to a maximum transfer rate allowed by the potentially congested bottleneck in the communication path.
21. The network of claim 18 wherein the control node is configured to rate limit the traffic stream by applying a feedback control process to modulate bandwidths in the traffic stream.
22. The network of claim 21 wherein the control node is upstream of the potentially congested bottleneck in the at least one communication path.
23. The network of claim 18 wherein the at least one communication path is selected on the basis of prior packet losses thereon.