Automatic discovery and enforcement of service level agreement settings

• US 8,917,596 B2
• Filed: 09/07/2012
• Issued: 12/23/2014
• Est. Priority Date: 09/07/2012
• Status: Active Grant

• Current Assignee: Accedian Networks Inc. (Cisco Systems, Inc.)
• Original Assignee: Accedian Networks Inc. (Cisco Systems, Inc.)
• Inventors: Rochon, Steve
• Primary Examiner(s): Nguyen, Hanh N
• Application Number:
  US13/606,135
• Publication Number:
  US 20140071822A1
• Time in Patent Office: 837 Days
• Field of Search: 370/241, 370/242, 370/468, 370/395.21, 370/392, 370/400, 370/352, 370/395.4, 370/230, 370/231, 370/232, 370/235, 370/236, 709/224, 709/225, 709/231
• US Class Current: 370/230
• CPC Class Codes: H04L 2012/6454 Random, e.g. Ethernet
• CPC Class Codes: H04L 41/0681 Configuration of triggering conditions
• CPC Class Codes: H04L 41/5009 Determining service level performance parameters or violations of service level contracts, e.g. violations of agreed response time or mean time between failures [MTBF]
• CPC Class Codes: H04L 41/5025 by proactively reacting to service quality change, e.g. by reconfiguration after service quality degradation or upgrade
• CPC Class Codes: H04L 41/5058 Service discovery by the service manager
• CPC Class Codes: H04L 43/024 by adaptive sampling
• CPC Class Codes: H04L 43/0829 Packet loss
• CPC Class Codes: H04L 43/0888 Throughput
• CPC Class Codes: H04L 47/21 using leaky-bucket
• CPC Class Codes: H04L 47/22 Traffic shaping
• CPC Class Codes: H04L 47/225 Determination of shaping rate, e.g. using a moving window
• CPC Class Codes: H04L 47/2425 for supporting services specification, e.g. SLA
• CPC Class Codes: H04L 47/29 using a combination of thresholds

Abstract

A method of automatically detecting the service level agreement settings in an Ethernet virtual circuit (EVC) carrying user packets comprises generating synthetic packets different from the user packets, on a data path in the EVC at a level sufficiently high to saturate the EVC. This accelerates the detection of actual effective values for the Committed Information Rate (CIR) and the Excess Information Rate (EIR) by triggering traffic shaping devices in the data path to enforce the CIR and EIR settings configured for those devices. A plurality of traffic samples are collected from the data path during a sampling period that is long enough to allow the collection of at least two valid traffic samples, where a valid traffic sample is the amount of user packets measured between two consecutive losses of synthetic packets. The CIR and EIR values enforced by the traffic shaping devices are determined from the collected samples.

Claims

1. A method of automatically detecting service level agreement (SLA) settings comprising committed information rate (CIR) and excess information rate (EIR) settings for a data path originating at a data source and terminating at a data sink, said data path carrying user packets and comprising traffic shaping devices configured with said CIR and EIR settings in accordance with said SLA, said method comprising

generating synthetic packets, different from the user packets, on said data path at a level that exceeds the CIR and EIR settings configured for the traffic shaping devices in said data path, said generating used to accelerate detection of actual values for CIR and EIR by triggering the traffic shaping devices in said data path to enforce, in accordance with said SLA, the CIR and EIR settings configured for said traffic shaping devices,

collecting a plurality of traffic samples from said data path during a sampling period that is long enough to allow the collection of at least two valid traffic samples, wherein each valid traffic sample corresponds to an amount of user packets measured between two consecutive losses of synthetic packets,

further wherein each of said valid traffic samples begins on a corresponding first packet loss indication and terminates on a corresponding second packet loss indication and includes a count of the total number of bytes received for the period between a pair of successive packet loss indications, and

determining from the collected traffic samples the actual CIR and EIR values enforced by said traffic shaping devices.

2. The method of claim 1 in which said CIR and EIR values are determined by analyzing each valid sample to determine a sum of the actual CIR and EIR based on the maximum number of consecutive data bytes within each valid traffic sample, and then subtracting, from the sum, the actual CIR based on the smallest number of data bytes in any of the collected samples, to obtain the actual EIR.

3. The method of claim 1 further comprising determining the value of the CIR and EIR settings.

4. The method of claim 1 in which said SLA settings are detected during service activation of an Ethernet service associated with said data path.

5. The method of claim 1 in which said SLA settings are detected during in-service operation of an Ethernet service associated with said data path.

6. The method of claim 1 in which said sampling period is bound by a starting sampling timestamp and an ending sampling timestamp.

7. The method of claim 6 in which the calculated traffic rate of a valid traffic sample is determined by dividing the total number of bytes received during said traffic sample by the difference between an end timestamp of said valid traffic sample and a begin timestamp of said valid traffic sample.

8. The method of claim 1 in which said count of the total number of bytes received is based on a sum of lengths in bytes of all user packets and synthetic packets received in said valid sample.

9. The method of claim 1 in which said synthetic packets include sequence numbers, and at least one of the corresponding first and second packet loss indications is determined by comparing the sequence number in a synthetic packet against an expected sequence number calculated as the value of the sequence number of the previous synthetic packet incremented by one.

10. The method of claim 1 in which said corresponding second packet loss indication is used as an end timestamp of a current traffic sample and where the same packet loss indication is used as a begin timestamp of the next traffic sample.

11. The method of claim 1 in which said valid traffic sample includes:

a begin timestamp taken after the starting sampling timestamp of said sampling period, and

an end timestamp taken before the end sampling timestamp of said sampling period.

12. The method of claim 1 in which said calculated values of the CIR and the EIR are obtained by:

setting the value of the CIR of said sampling period to zero,

setting the value of the EIR of said sampling period to zero, and

analyzing the collected valid traffic samples obtained for the said sampling period.

13. The method of claim 1 in which the calculated values of the CIR and the EIR for said sampling period are obtained by:

determining whether the calculated value of the CIR is equal to zero and if it is, setting the value of the CIR to the value of the calculated traffic rate for the current traffic sample being analyzed;

otherwise, determining if the calculated traffic rate for said traffic sample is less than the calculated value of the CIR and if it is, calculating the effective value of the EIR by adding the value of the CIR to the value of the EIR and deducting the value of the calculated traffic rate for the current traffic sample being analyzed, and setting the value of the CIR to the value of the calculated traffic rate for said traffic sample;

otherwise, if the calculated traffic rate for said traffic sample is larger than the sum of the effective CIR and of the EIR, adjusting the effective value of the EIR for the said sampling period to the difference between the effective data rate for said sample and the effective CIR for said sampling period.

14. A method of automatically detecting service level agreement (SLA) settings comprising committed information rate (CIR) and excess information rate (EIR) settings for a data path originating at a data source and terminating at a data sink, said data path carrying user packets and comprising traffic shaping devices configured with said CIR and EIR settings in accordance with said SLA, said method comprising

generating synthetic packets, different from the user packets, on said data path at a level that exceeds the CIR and EIR settings configured for the traffic shaping devices in said data path, said generating used to accelerate detection of actual values for CIR and EIR by triggering the traffic shaping devices in said data path to enforce, in accordance with said SLA, the CIR and EIR settings configured for said traffic shaping devices,

collecting a plurality of traffic samples from said data path during a sampling period that is long enough to allow the collection of at least two valid traffic samples, wherein each valid traffic sample corresponds to an amount of user packets measured between two consecutive losses of synthetic packets, where said sampling period is bound by a starting sampling timestamp and an ending sampling timestamp;

and

determining the calculated traffic rate of a valid traffic sample by dividing the total number of bytes received during the said traffic sample by the difference between an end timestamp of said valid traffic sample and a begin timestamp of said valid traffic sample.

15. The method of claim 14 in which said CIR and EIR values are determined by analyzing each valid sample to determine a sum of the actual CIR and EIR based on a maximum number of consecutive data bytes in each valid traffic sample, and then subtracting, from the sum, the actual CIR based on the smallest number of data bytes in any of the collected samples, to obtain the actual EIR.

16. The method of claim 14 in which

said begin timestamp is taken after the starting sampling timestamp of said sampling period, and

said end timestamp is taken before the ending sampling timestamp of said sampling period.

17. The method of claim 14 in which said SLA settings are detected during service activation of an Ethernet service associated with said data path.

18. The method of claim 14 in which said SLA settings are detected during in-service operation of an Ethernet service associated with said data path.

19. A method of automatically detecting service level agreement (SLA) settings comprising committed information rate (CIR) and excess information rate (EIR) settings for a data path originating at a data source and terminating at a data sink, said data path carrying user packets and comprising traffic shaping devices configured with said CIR and EIR settings in accordance with said SLA, said method comprising

generating synthetic packets, different from the user packets, on said data path at a level that exceeds the CIR and EIR settings configured for the traffic shaping devices in said data path, said generating used to accelerate detection of actual values for CIR and EIR by triggering the traffic shaping devices in said data path to enforce, in accordance with said SLA, the CIR and EIR settings configured for said traffic shaping devices,

collecting a plurality of traffic samples from said data path during a sampling period that is long enough to allow the collection of at least two valid traffic samples, wherein each valid traffic sample corresponds to an amount of user packets measured between two consecutive losses of synthetic packets, where said sampling period is bound by a starting sampling timestamp and an ending sampling timestamp;

and

determining from the collected traffic samples the actual CIR and EIR values enforced by said traffic shaping devices, wherein said determining is performed by analyzing each valid sample to determine a sum of the actual CIR and EIR based on a maximum number of consecutive data bytes in each valid traffic sample, and then subtracting, from the sum, the actual CIR based on the smallest number of data bytes in any of the collected samples, to obtain the actual EIR.

20. The method of claim 19 further comprising determining the value of the CIR and EIR settings.