Selective compression for network connections

• US 8,010,668 B1
• Filed: 12/29/2010
• Issued: 08/30/2011
• Est. Priority Date: 10/01/2004
• Status: Active Grant

• Current Assignee: F5 Networks, Inc. (F5 Incorporated)
• Original Assignee: F5 Networks, Inc. (F5 Incorporated)
• Inventors: Rothstein, Jesse A., Mukerji, Arindum
• Primary Examiner(s): Dennison; Jerry
• Application Number:
  US12/981,338
• Time in Patent Office: 244 Days
• Field of Search: 709/224, 709/249
• US Class Current: 709/224
• CPC Class Codes: H04L 69/04 Protocols for data compression, e.g. ROHC
• CPC Class Codes: H04W 4/18 Information format or content conversion, e.g. adaptation by the network of the transmitted or received information for the purpose of wireless delivery to users or terminals

Abstract

A system, apparatus, and method selectively provides content compression to a client based, in part, on whether the network connection from the client is determined to be a high latency, low-bandwidth connection. The present invention gathers one or more network metrics associated with the connection from the client. In one embodiment, the metrics include estimated TCP metrics, including smoothed round trip time, maximum segment size (MSS), and bandwidth delay product (BWDP). These estimated network metrics are employed to make an application layer decision of whether the client connection is a high latency, low-bandwidth connection. If it is, then content may be selectively compressed virtually on the fly for transfer over the network connection. In one embodiment, the selective compression uses a content encoding compression feature of the HTTP protocol standard.

Claims

1. A network device for managing a communication with a client device over a network, comprising:

a transceiver for receiving and sending information to the client device through at least one of a plurality of gateways being interposed between the client device and at least a first and a second target server; and

a traffic management component operating on the network device and that is configured to perform actions, including;

determining a plurality of network metrics for at least one client device/gateway combination;

combining results of evaluations of each of the plurality of network metrics to a different threshold value into a single network connection characteristic for the at least one client device/gateway combination;

selecting from the first or the second target server, based at least in part on the single network connection characteristic, wherein the first target server storing low bandwidth content is selected for network connections having a low bandwidth network connection characteristic, and the second target server storing high bandwidth content is selected for network connections having a high bandwidth network connection characteristic; and

enabling a communication between the client device and the selected first or second target server.

2. The network device of claim 1, wherein traffic management component is configured to perform actions further comprising selectively enabling compression of content that is sent towards the client device, based on the single network connection characteristic.

3. The network device of claim 1, wherein the plurality of network metrics includes at least two of a round trip time (RTT), maximum segment size (MSS), and a bandwidth delay product (BWDP).

4. The network device of claim 3, wherein combining results of evaluations further comprises:

comparing each of the at least two network metrics to a different threshold value;

obtaining a result for each of the comparisons; and

combining the results to determine the single network connection characteristic.

5. The network device of claim 1, wherein the single network connection characteristic indicates whether the at least one client device/gateway combination is a high latency and low bandwidth network connection.

6. The network device of claim 1, wherein determining a plurality of network metrics further comprises determining that at least one of the network metrics is unavailable for client device/gateway combination, and in response estimating the unavailable network metric.

7. A system, comprising:

a first server device and a second server device, each server device being configured to store content; and

a network device interposed between a client device and the first and second server devices, the network device being configured to perform actions, including;

determining a plurality of network metrics for at least one client device/gateway combination;

combining results of evaluations of each of the plurality of network metrics to a different threshold value into a single network connection characteristic for the at least one client device/gateway combination;

determining whether the combined results indicate that a network connection to the at least one client device/gateway combination is a low bandwidth high latency connection; and

based on the determination, selecting from the first or the second server device, wherein the first server device is configured to provide content for at least low bandwidth high latency connections; and

enabling a communication between the client device and the selected first or second server device.

8. The system of claim 7, wherein the first server device is configured to provide content that is currently compressed content for low bandwidth high latency connections.

9. The system of claim 7, wherein the first server device is configured to compress content for low bandwidth high latency connections prior to providing the content towards the at least one client device/gateway combination.

10. The system of claim 7, wherein determining a plurality of network metrics further comprises when it is determined that at least one of the network metrics is unavailable for the current client device/gateway combination, estimating the unavailable network metric based on a previous value of the network metric determined from a previous connection for the same client device/gateway combination.

11. The system of claim 7, wherein the plurality of network metrics includes at least two of a round trip time (RTT), maximum segment size (MSS), and a bandwidth delay product (BWDP).

12. The system of claim 7, wherein at least one network metric is a maximum segment size (MSS) that is initialized based on a value supplied during an initial synchronization from the client device.

13. The system of claim 7, wherein the network metrics are obtained during the client device/gateway connection over time as an estimate determined from a moving average.

14. A method operating on one or more network devices, comprising:

determining a plurality of network metrics for at least one client device/gateway combination, the one or more network devices being interposed between a client device and at least one gateway and a first server device and a second server device;

combining results of evaluations of each of the plurality of network metrics to a different threshold value into a single network connection characteristic for the at least one client device/gateway combination;

determining whether the combined results indicate that a network connection to the at least one client device/gateway combination is a low bandwidth high latency connection; and

based on the determination, selecting from the first or the second server device, wherein the first server device is configured to provide content for at least low bandwidth high latency connections; and

enabling a communication between the client device and the selected first or second server device.

15. The method of claim 14, wherein the first server device is configured to compress content for low bandwidth high latency connections prior to providing the content towards the at least one client device/gateway combination.

16. The method of claim 14, wherein determining a plurality of network metrics further comprises when it is determined that at least one of the network metrics is unavailable for the current client device/gateway combination, estimating the unavailable network metric based on a previous value of the network metric determined from a previous connection for the same client device/gateway combination.

17. The method of claim 14, wherein the plurality of network metrics includes at least two of a round trip time (RTT), maximum segment size (MSS), and a bandwidth delay product (BWDP).

18. The method of claim 14, wherein at least one network metric is a maximum segment size (MSS) that is initialized based on a value supplied during an initial synchronization from the client device.

19. The method of claim 14, wherein the network metrics are obtained during the client device/gateway connection over time as an estimate determined from a moving average.

20. The method of claim 14, wherein the plurality of network metrics are refined over time, such that determining whether the combined results indicate that a network connection for the at least one client device/gateway combination is a low bandwidth high latency connection is revised over time.