Detecting a lack of correlation between application quality of experience and network conditions

1. A method comprising:

obtaining, by a device, quality of experience metrics for an online application;

obtaining, by the device, network metrics for one or more network paths over which traffic for the online application was routed;

identifying, by the device, a lack of correlation between the quality of experience metrics for the online application and the network metrics for the one or more network paths over which traffic for the online application was routed; and

disabling, by the device and based on the lack of correlation, explicit probing of the one or more network paths.

2. The method as in claim 1 , wherein the one or more network paths are part of a software defined wide area network (SD-WAN).

3. The method as in claim 1 , further comprising:

deactivating, by the device, a predictive routing engine from making routing decisions for the traffic for the online application, wherein the predictive routing engine uses as input results of the explicit probing of the one or more network paths.

4. The method as in claim 3 , wherein the predictive routing engine uses machine learning to predict service level agreement (SLA) violations for the online application by the one or more network paths.

5. The method as in claim 1 , further comprising:

determining whether the quality of experience metrics for the online application are correlated with network metrics for one or more new paths to which the traffic for the online application has been moved.

6. The method as in claim 5 , further comprising:

identifying movement of the online application to a new address, using Domain Name System (DNS) information associated with the online application.

7. The method as in claim 1 , wherein identifying the lack of correlation comprises:

computing one or more correlation scores between the quality of experience metrics and the network metrics; and

determining whether the one or more correlation scores exceed a predefined threshold.

8. The method as in claim 1 , wherein the network metrics comprise at least one of: a round trip time for the traffic, an amount of jitter for the traffic, a loss percentage for the traffic, a bitrate for the traffic, or a recovery percentage for the traffic.

9. The method as in claim 1 , further comprising:

providing an indication of the lack of correlation to a user interface.

10. The method as in claim 1 , wherein the device identifies the lack of correlation based on a request to assess a behavior of the online application or the one or more network paths.

11. An apparatus, comprising:

one or more network interfaces;

a processor coupled to the one or more network interfaces and configured to execute one or more processes; and

a memory configured to store a process that is executable by the processor, the process when executed configured to:

obtain quality of experience metrics for an online application;

obtain network metrics for one or more network paths over which traffic for the online application was routed;

identify a lack of correlation between the quality of experience metrics for the online application and the network metrics for the one or more network paths over which traffic for the online application was routed; and

disable, based on the lack of correlation, explicit probing of the one or more network paths.

12. The apparatus as in claim 11 , wherein the one or more network paths are part of a software defined wide area network (SD-WAN).

13. The apparatus as in claim 11 , wherein the process when executed is further configured to:

deactivate a predictive routing engine from making routing decisions for the traffic for the online application, wherein the predictive routing engine uses as input results of the explicit probing of the one or more network paths.

14. The apparatus as in claim 13 , wherein the predictive routing engine uses machine learning to predict service level agreement (SLA) violations for the online application by the one or more network paths.

15. The apparatus as in claim 11 , wherein the process when executed is further configured to:

determine whether the quality of experience metrics for the online application are correlated with network metrics for one or more new paths to which the traffic for the online application has been moved.

16. The apparatus as in claim 15 , wherein the process when executed is further configured to:

identify movement of the online application to a new address, using Domain Name System (DNS) information associated with the online application.

17. The apparatus as in claim 11 , wherein the apparatus identifies the lack of correlation by:

computing one or more correlation scores between the quality of experience metrics and the network metrics; and

determining whether the one or more correlation scores exceed a predefined threshold.

18. The apparatus as in claim 11 , wherein the network metrics comprise at least one of: a round trip time for the traffic, an amount of jitter for the traffic, a loss percentage for the traffic, a bitrate for the traffic, or a recovery percentage for the traffic.

19. The apparatus as in claim 11 , wherein the process when executed is further configured to:

provide an indication of the lack of correlation to a user interface.

20. A tangible, non-transitory, computer-readable medium storing program instructions that cause a device to execute a process comprising:

obtaining, by the device, quality of experience metrics for an online application;

obtaining, by the device, network metrics for one or more network paths over which traffic for the online application was routed;

identifying, by the device, a lack of correlation between the quality of experience metrics for the online application and the network metrics for the one or more network paths over which traffic for the online application was routed; and

disabling, by the device and based on the lack of correlation, explicit probing of the one or more network paths.