Automatic framing selection

• US 20070064717A1
• Filed: 07/26/2006
• Published: 03/22/2007
• Est. Priority Date: 07/28/2005
• Status: Active Grant

• Current Assignee: Riverbed Technology, LLC (Riverbed Technology Incorporated)
• Original Assignee: Riverbed Technology Incorporated
• Inventors: Gupta, Nitin, Keswani, Vivasvat, Subbana, Kartik, Tze-Si Wu, David, Lasen, Soren
• Granted Patent: US 8,204,058 B2
• Time in Patent Office: Days
• US Class Current: 370/412
• CPC Class Codes: H04L 12/28 characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks] wireless communication networks H04W ; arrangements for dividing the transmission path H04W40/00
• CPC Class Codes: H04L 43/16 Threshold monitoring
• CPC Class Codes: H04L 47/26 using explicit feedback to the source, e.g. choke packets
• CPC Class Codes: H04L 47/27 Evaluation or update of window size, e.g. using information derived from acknowledged [ACK] packets
• CPC Class Codes: H04L 69/16 Implementation or adaptation of Internet protocol [IP], of transmission control protocol [TCP] or of user datagram protocol [UDP]
• CPC Class Codes: H04L 69/163 In-band adaptation of TCP data exchange; In-band control procedures

Abstract

Network traffic is monitored and an optimal framing heuristic is automatically determined and applied. Framing heuristics specify different rules for framing network traffic. While a framing heuristic is applied to the network traffic, alternative framing heuristics are speculatively evaluated for the network traffic. The results of these evaluations are used to rank the framing heuristics. The framing heuristic with the best rank is selected for framing subsequent network traffic. Each client/server traffic flow may have a separate framing heuristic. The framing heuristics may be deterministic based on byte count and/or time or based on traffic characteristics that indicate a plausible point for framing to occur. The choice of available framing heuristics may be determined partly by manual configuration, which specifies which framing heuristics are available, and partly by automatic processes, which determine the best framing heuristic to apply to the current network traffic from the set of available framing heuristics.

Claims

1. A method of determining an optimal framing of network traffic, the method comprising:

receiving a data stream over a time period;

evaluating a plurality of framing heuristics on the data stream;

ranking at least a portion of the plurality of framing heuristics according to their behavior during the time period; and

selecting at least one of the plurality of framing heuristics in response to the ranking.

2. The method of claim 1, further comprising:

providing the selected ones of the plurality of framing heuristics to a controller; and

framing network traffic using the selected ones of the plurality of framing heuristics.

3. The method of claim 2, wherein the network traffic includes the data stream.

4. The method of claim 1, wherein the time period is determined by a selected framing heuristic applied to the data stream.

5. The method of claim 4, wherein the selected framing heuristic implements a default framing behavior.

6. The method of claim 5, wherein the default framing behavior limits the time period to the lesser of a predetermined elapsed time or a predetermined size of the received data stream.

7. The method of claim 4, wherein the selected framing heuristic is determined from a previous iteration of receiving the data stream, evaluating the plurality of framing heuristics on the data stream, ranking at least a portion of the plurality of framing heuristics, and selecting at least one of the plurality of framing heuristics.

8. The method of claim 1, wherein:

evaluating the plurality of framing heuristics comprises, for each of the plurality of framing heuristics, activating the framing heuristic in response to the data stream satisfying to the framing heuristic; and

ranking at least a portion of the plurality of framing heuristics comprises determining a scoring metric value for at least the activated framing heuristics.

9. The method of claim 8, wherein:

activating the framing heuristic further comprises determining for each activated framing heuristic a size of the data stream at the time the framing heuristic is satisfied; and

determining a scoring metric value comprises;

comparing the sizes associated with the activated framing heuristics to determine a size rank value for each activated framing heuristic; and

assigning the scoring metric values to the activated framing heuristics according to their respective size rank values.

10. The method of claim 8, wherein selecting at least one of the plurality of framing heuristics in response to the ranking comprises:

selecting at least one of the plurality of framing heuristics having a scoring metric value better than a threshold value.

11. The method of claim 8, wherein selecting at least one of the plurality of framing heuristics in response to the ranking comprises:

selecting at least one of the plurality of framing heuristics having a scoring metric value better than a scoring metric value of a currently selected framing heuristic, wherein the scoring metric value of the currently selected framing heuristic is determined by evaluating the currently selected framing heuristic on the data stream.

12. The method of claim 1, wherein receiving the data stream comprises storing the data stream in a data buffer and wherein the plurality of framing heuristics are adapted to trigger a flush of the data buffer, thereby providing data for at least one output network packet.

13. A method of framing network traffic, the method comprising:

receiving a first input packet data;

adding the first input packet data to buffered data;

in response to receiving the first input packet data, evaluating a plurality of framing heuristics on at least a portion of the buffered data;

for each of the plurality framing heuristics, activating the framing heuristic in response to the portion of the buffered data satisfying the framing heuristic;

updating scoring metric values for at least the activated framing heuristics in response to the determination that the portion of the buffered data satisfies a selected one of the plurality of framing heuristics;

initiating a flush of the buffered data in response to the determination that the portion of the buffered data satisfies the selected one of the plurality of framing heuristics; and

selecting at least a different one of the plurality of framing heuristics in response to the updated scoring metric satisfying a criteria.

14. The method of claim 13, further comprising repeating the steps of receiving, adding, evaluating, activating for a second input packet data.

15. The method of claim 13, wherein the criteria includes the different one of the plurality of framing heuristics having a scoring metric value better than a threshold value.

16. The method of claim 13, wherein the criteria includes the different one of the plurality of framing heuristics having a scoring metric value better than the scoring metric value of the selected one of the plurality of framing heuristics.

17. The method of claim 13, further comprising:

in response to the portion of the buffered data not satisfying the selected one of the plurality of framing heuristics, initiating a flush of the buffered data in response to the determination that a default framing heuristic is satisfied.

18. The method of claim 17, wherein the default framing heuristic is adapted to initiate the flush of the buffered data in response to an expiration of a predetermined time period or a size of the buffered data exceeding a predetermined size limit.

19. The method of claim 13, further comprising:

outputting at least one output packet including at least a portion of the buffered data.

20. The method of claim 13, wherein updating scoring metric values for at least the activated framing heuristics comprises:

increasing a scoring metric value of at least a portion of the activated framing heuristics.

21. The method of claim 20, wherein activating the framing heuristic comprises storing a size of the buffered data at the time the portion of the buffered data satisfies the framing heuristic;

and

wherein the increase of the scoring metric value for each of the portion of the activated framing metrics is based on the respective size of the buffered data at the time the portion of the buffered data satisfies the framing heuristic.

22. The method of claim 13, wherein at least one of the plurality of framing heuristics is adapted to be activated in response to a TCP flag.

23. The method of claim 13, wherein at least one of the plurality of framing heuristics is adapted to be activated in response to the size of the buffered data being equal to a MTU size.

24. The method of claim 13, wherein at least one of the plurality of framing heuristics is adapted to be activated in response to carriage return/line feed character.

25. The method of claim 13, wherein at least one of the plurality of framing heuristics is adapted to be activated in response to at least one character delimiting a transaction.

26. The method of claim 13, wherein at least one of the plurality of framing heuristics is adapted to be activated in response to an SQL transaction.