Method and system for resource-aware dynamic bandwidth control

• US 8,667,166 B2
• Filed: 11/02/2010
• Issued: 03/04/2014
• Est. Priority Date: 11/02/2010
• Status: Active Grant

• Current Assignee: Wickr Inc. (Amazon.com, Inc.)
• Original Assignee: Net Power & Light, Inc (Amazon.com, Inc.)
• Inventors: Dubovik, Gleb, Shtayura, Vadim, Surin, Nikolay
• Primary Examiner(s): VU, VIET DUY
• Application Number:
  US12/938,249
• Publication Number:
  US 20120110162A1
• Time in Patent Office: 1,218 Days
• Field of Search: 709/217, 709/219, 709/223, 709/224, 709/226, 709/227, 709/229, 709/250, 709/230, 709/232, 725/105, 725/109, 725/118, 725/120
• US Class Current: 709/232
• CPC Class Codes: H04L 12/6418 Hybrid transport
• CPC Class Codes: H04L 41/0896 Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities flow or congestion control using dynamic resource allocation, e.g. in-call renegotiation, H04L47/76
• CPC Class Codes: H04L 47/11 Identifying congestion
• CPC Class Codes: H04L 47/12 Avoiding congestion; Recovering from congestion
• CPC Class Codes: H04L 47/25 with rate being modified by the source upon detecting a change of network conditions
• CPC Class Codes: H04L 47/263 Rate modification at the source after receiving feedback
• CPC Class Codes: H04L 65/70 Media network packetisation

Abstract

Resource-aware dynamic bandwidth control uses information about current network state and receiver performance to avoid, minimize and/or recover from the effects of network spikes and data processing spikes. Linear models may be used to estimate a time required to process data packets in a data processing queue, and are thus useful to determine whether a data processing spike is occurring. When a data processing spike occurs, an alarm may be sent from a client to a server notifying the server that the client must drop packets. In response, the server can encode and transmit an independent packet suitable for replacing the queued data packets which can then be dropped by the client and the independent packet present to the processor instead.

Claims

1. A computer implemented method

transmitting a specific data stream across a network from a first network node to a second network node according to a given output bandwidth;

monitoring a network state of the network;

monitoring an input bandwidth of the second network node including;

receiving initial data packets corresponding to the specific data stream at the second network node;

inserting the initial data packets into a processing queue;

measuring a latency characteristic of the processing queue;

when the latency characteristic of the processing queue indicates that the initial data packets cannot be processed in a timely manner such that a processing task cannot be timely performed, sending an alarm signal from the second network node to the first network node;

dynamically adjusting the output bandwidth according to the network state and the input bandwidth such that the processing task for the second network node indicated by the specific data stream can be performed; and

upon receiving the alarm signal at the first network node, sending a specific packet from the first network node to the second network node, the specific packet intended to replace one or more of the initial data packets and enable the second network node to accomplish the processing task.

2. A computer implemented method as recited in claim 1, wherein transmitting specific data across the network is performed according to a low-latency protocol.

3. A computer implemented method as recited in claim 1, wherein monitoring the network state includes determining a network latency characteristic of the network.

4. A computer implemented method as recited in claim 3, wherein determining the network latency characteristic of the network is accomplished by transmitting test packets.

5. A computer implemented method as recited in claim 1, wherein the input bandwidth of the second network node is associated with a data processing rate of the second network node, the data processing rate of the second network node associated with processing the specific data received at the second network node.

6. A computer implemented method as recited in claim 5, wherein the output bandwidth corresponds to a specific bitrate.

7. A computer implemented method as recited in claim 6, wherein dynamically adjusting the output bandwidth includes raising or lowering the specific bitrate.

8. A computer implemented method as recited in claim 7, wherein lowering the specific bitrate includes increasing a compression factor when encoding the specific data stream.

9. A computer implemented method as recited in claim 1, further comprising, upon receiving the specific packet at the second network node, dropping any remaining initial packets from the processing queue and replacing the dropped initial packets with the specific packet.

10. A computer implemented method as recited in claim 9, further comprising lowering the output bandwidth in order to decrease the processing queue latency.

11. A system for providing resource-aware dynamic bandwidth control comprising:

a server network node;

a client network node coupled to the server network node via a communications network;

wherein, the system is operable to monitor state information of the communications network, and state information of a data processor associated with the client network node;

wherein, the server network node is operable to transmit data packets to the client network node according to an output bandwidth, and to dynamically adjust the output bandwidth based on state information of the communications network, and state information of the data processor associated with the client network node;

wherein, the client network node is operable to process data packets received from the server network node according to an input bandwidth, wherein the input bandwidth corresponds to a data packet processing rate of the data processor associated with the client network node;

wherein, the client network node is operable to;

insert the data packets into a processing queue;

measure a latency characteristic of the processing queue;

when the latency characteristic of the processing queue indicates that the initial data packets cannot be processed in a timely manner such that a processing task cannot be timely performed, send an alarm signal to the server network node;

wherein the client network node is further operable to identify a data packet processing spike;

wherein the client network node is operable to send an alarm to the server network node responsive to identifying the data packet processing spike; and

wherein the server network node is responsive to the alarm to generate an independent data packet corresponding to a workload indicated by data packets previously transmitted to the client network node, wherein the independent data packet can be processed independent of any other data packets.

12. A system as recited in claim 11, wherein the server network node transmits the independent data packet to the client network node, and the client network node discards previously received data packets that have not been processed, and presents the independent data packet to the data processor for operation.

16. A computer implemented method as recited in claim 12, wherein dynamically adjusting the output bandwidth includes raising or lowering the specific bitrate.

17. A computer implemented method as recited in claim 16, wherein lowering the specific bitrate includes increasing a compression factor when encoding the specific data stream.

13. A computer implemented method comprising:

transmitting a specific data stream across a network from a first network node to a second network node according to a given output bandwidth;

monitoring an input bandwidth of the second network node, wherein the input bandwidth of the second network node is associated with a data processing rate of the second network node, the data processing rate of the second network node associated with processing the specific data received at the second network node;

wherein monitoring the input bandwidth including;

receiving initial data packets corresponding to the specific data stream at the second network node;

inserting the initial data packets into a processing queue;

measuring a latency characteristic of the processing queue;

when the latency characteristic of the processing queue indicates that the initial data packets cannot be processed in a timely manner such that a processing task cannot be timely performed, sending an alarm signal from the second network node to the first network node;

dynamically adjusting the output bandwidth with reference to the input bandwidth such that the processing task for the second network node indicated by the specific data stream can be performed; and

upon receiving the alarm signal at the first network node, sending a specific packet from the first network node to the second network node, the specific packet intended to replace one or more of the initial data packets and enable the second network node to accomplish the processing task.

14. A computer implemented method as recited in claim 13, wherein transmitting the specific data stream across the network is performed according to a low-latency protocol.

15. A computer implemented method as recited in claim 13, wherein the output bandwidth corresponds to a specific bitrate.

18. A computer implemented method as recited in claim 13, further comprising:

lowering the output bandwidth of the first network node in order to decrease the latency characteristic the processing queue.