Metacodec for optimizing network data compression based on comparison of write and read rates

• US 8,417,833 B1
• Filed: 11/29/2006
• Issued: 04/09/2013
• Est. Priority Date: 11/29/2006
• Status: Active Grant

• Current Assignee: F5 Networks, Inc. (F5 Incorporated)
• Original Assignee: F5 Networks, Inc. (F5 Incorporated)
• Inventors: Amdahl, Saxon Carl
• Primary Examiner(s): Boutah, Alina N
• Assistant Examiner(s): John, Clarence
• Application Number:
  US11/564,805
• Time in Patent Office: 2,323 Days
• Field of Search: 370/389, 709/247, 711167-173
• US Class Current: 709/247
• CPC Class Codes: H04L 69/04 Protocols for data compression, e.g. ROHC

Abstract

A method, system, and apparatus are directed towards selectively compressing data for transmission over a network. In one embodiment, a sending network device and receiving network device negotiates different compression modes to communicate data between them. An initial compression mode may be selected based on a network bandwidth. The sending network device then reads data, and compresses using the selected compression mode. The compressed data may then be written out. Ratios of compression and the write times are then employed to selectively adjust the compression mode for subsequent data compressions. In one embodiment, a compression ratio is also employed to determine whether to employ the selected compression mode, or to reduce the level of compression by using a different compression mode. The receiving network device having received information about the selected compression mode, then employs that compression mode to decompress the received data.

Claims

1. A network device that manages data over a network, comprising:

a network communication component that sends or receives data over the network; and

a processor that performs actions involving the data, including;

reading the data;

compressing the data using a compression mode;

determining a compression time for compressing the data;

writing the compressed data to a buffer for use in transmitting the compressed data over the network;

determining a write time for writing the compressed data to the buffer;

determining a ratio by dividing the compression time by the write time; and

selectively modifying the compression mode based in part on the determined ratio of compression time divided by write, wherein selectively modifying the compression mode further comprises;

if the ratio of compression time/write time is greater than a threshold value, selectively selecting another compression mode having a lower expected compression ration than a currently selected compression mode; and

if a ratio of write time/compression time is greater than another threshold value, selectively selecting yet another compression mode having a higher expected compression ration than a currently selected compression mode.

2. The network device of claim 1, wherein the processor performs actions, further including:

determining a compression ratio, in part, from a number of bits in the read data and a number of bits in the compressed data;

reading another data; and

if the compression ratio indicates that the data is incompressible, employing a reduced compression mode over a currently selected compression mode to attempt to compress the other data.

3. The network device of claim 2, wherein employing the reduced compression mode further comprises selecting a compression mode from a plurality of available compression modes with a fastest computation time.

4. The network device of claim 1, wherein selectively selecting another compression mode further comprises employing a detent counter to minimize dithering between compression modes.

5. The network device of claim 1, wherein the compression mode is selectable from a plurality of compression modes, each of the plurality being configured to provide a different expected compression ratios or expected execution times.

6. The network device of claim 1, wherein compressing the data using a compression mode further comprises selecting an initial compression mode based, in part, on a network bandwidth characteristic.

7. The network device of claim 1, wherein selectively modifying the compression mode further comprises negotiating with a receiving device at least one compression mode.

8. The network device of claim 1, wherein writing the compressed data to an output buffer further comprises, writing the compressed data to at least one of a TCP stack buffer, or a network card buffer.

9. A system configured to communicate data over a network, comprising:

a first network device that is operative to perform actions, including;

selecting a first compression mode from a plurality of compression modes;

compressing a first portion of data using the first compression mode;

determining a compression time for compressing the first portion of data;

writing the compressed first portion of the data for use in transmitting the compressed first portion of the data over the network;

determining a write time for writing the compressed first portion of data;

determining a ratio by dividing the compression time by the write time;

selectively selecting a second compression mode based in part on the determined compression time divided by write time ratio, wherein selectively selecting the second compression mode further comprises;

if the ratio of compression time/write time is greater than a threshold value, selectively reducing the first compression mode; and

if a ratio of write time/compression time is greater than another threshold value, selectively increasing the first compression mode; and

compressing a second portion of the data using the second compression mode; and

a second network device that is operative to perform actions, including;

negotiating with the first network device the plurality of compression modes;

receiving the compressed data;

determining one of the plurality of compression modes used to compress the data; and

decompressing the compressed data using the determined compression mode.

10. The system of claim 9, wherein the first network device is operative to perform actions, further comprising:

determining a compression ratio using a weighted running average of compression ratios over time; and

employing the compression ratio to selectively select a reduced compression mode for another data.

11. The system of claim 9, wherein the second network device receives information about a compression mode employed to compress at least a portion of the data in data header of a network packet.

12. The system of claim 9, wherein selectively reducing or increasing the first compression mode further comprises employing a detent counter to minimize dithering between compression modes, the detent counter being further weighted based on a number of bits in the data to be compressed.

13. The system of claim 9, wherein the first network device is operative to perform actions, further comprising:

determining a compression ratio; and

if the compression ratio indicates that the data is incompressible, selecting the second compression mode based, in part, as a fastest computation time compression mode from a plurality of available compression modes.

14. The system of claim 13, wherein the fastest computation time compression mode further comprises a compression mode with zero compression.

15. A method of managing content over a network, comprising;

compressing a first portion of data using a first compression mode;

determining a compression time for compressing the first portion of the data;

writing out the compressed first portion of the data to a buffer;

determining a write time for writing the compressed first portion of the data to the buffer;

determining a ratio by dividing the compression time by the write time; and

selectively selecting a second compression mode based in part on the determined ratio and compressing a second portion of the data, wherein selectively selecting the second compression mode further comprises;

if the ratio of compression time/write time is greater than a threshold value, selectively reducing the first compression mode; and

if a ratio of write time/compression time is greater than another threshold value, selectively increasing the first compression mode.

16. The method of claim 15, further comprising:

negotiating at least one compression mode; and

selecting the first compression mode based, in part, on a network connection characteristic.

17. The method of claim 15, wherein selectively selecting a second compression mode further comprises:

if the ratio of compression time divided by write time is greater than the threshold value, incrementing a tick value, and

if the tick value also exceeds yet another threshold value, selecting the second compression mode as having an expected lower compression ratio over that of the first compression mode.

18. The method of claim 15, wherein selectively selecting a second compression mode further comprises:

if a ratio of write time divided by compression time is greater than the other threshold value, incrementing a tick value, and

if the tick value also exceeds yet another threshold value, selecting the second compression mode as having an expected higher compression ratio over that of the first compression mode.

19. The method of claim 15, wherein the tick value is further weighted by a number of bits in the first portion of uncompressed data.

20. A non-transitory, computer-readable storage medium that stores computer-executable program instructions when installed on a network device performs the method of claim 15.

21. A traffic management device (TMD) for enabling communications between network devices, comprising:

a transceiver for sending and receiving network packets; and

a metacodec component operative to perform actions, including;

reading data from another computing device;

compressing the data using a first compression mode;

calculating a compression time for compressing the data;

writing the compressed data to a buffer for use in transmitting the compressed data over the network;

calculating a write time for writing the compressed data to the buffer;

determining a ratio by dividing the compression time by the write time;

selectively selecting a second compression mode based in part on the determined ratio of compression time to write time, wherein selectively selecting the second compression mode further comprises;

if the ratio of compression time/write time is greater than a threshold value, selectively reducing the first compression mode; and

if a ratio of write time/compression time is greater than another threshold value, selectively increasing the first compression mode;

reading additional data from the other computing device; and

employing the second compression mode to selectively compress the additional data.