Detecting application performance breaking points based on uncertainty and active learning
In one embodiment, a device obtains path metrics for a network path via which traffic for an online application is conveyed. The device models uncertainty of an application experience metric predicted for the online application based on the path metrics. The device identifies, based on the uncertainty of the application experience metric modeled by the device, a breaking point in the path metrics at which the application experience metric predicted for the online application is expected to switch from being acceptable to being degraded. The device provides the breaking point in the path metrics for display.
1. A method comprising:
obtaining, by a device, path metrics for a network path via which traffic for an online application is conveyed;
modeling, by the device, uncertainty of an application experience metric predicted for the online application based on the path metrics;
identifying, by the device and based on the uncertainty of the application experience metric modeled by the device, a breaking point in the path metrics at which the application experience metric predicted for the online application is expected to switch from being acceptable to being degraded, wherein the breaking point in the path metrics is located in a region of higher uncertainty of the application experience metric than a region where the application experience metric is predicted to be acceptable for the online application; and
providing, by the device, the breaking point in the path metrics for display.
2. The method as in claim 1 , wherein the application experience metric comprises a service level agreement associated with the network path.
3. The method as in claim 1 , wherein the path metrics comprise at least one of: loss, jitter, delay, throughput, or metrics from a particular device associated with the network path.
4. The method as in claim 1 , wherein identifying the breaking point in the path metrics comprises:
causing performance of an experiment along the network path, to obtain one or more samples from the network path, wherein the breaking point is identified based in part on those one or more samples.
5. The method as in claim 4 , further comprising:
recommending the experiment to a user interface, wherein the device causes performance of the experiment in response to an instruction to do so from the user interface.
6. The method as in claim 1 , further comprising:
initiating, by the device and based on the breaking point, a reroute of at least a portion of the traffic of the online application.
7. The method as in claim 1 , wherein the application experience metric is predicted by a machine learning model for the online application.
8. The method as in claim 1 , wherein the online application is a software-as-a-service (SaaS) application.
9. 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 path metrics for a network path via which traffic for an online application is conveyed;
model an uncertainty of an application experience metric predicted for the online application based on the path metrics;
identify, based on the uncertainty of the application experience metric modeled by the apparatus, a breaking point in the path metrics at which the application experience metric predicted for the online application is expected to switch from being acceptable to being degraded, wherein the breaking point in the path metrics is located in a region of higher uncertainty of the application experience metric than a region where the application experience metric is predicted to be acceptable for the online application; and
provide the breaking point in the path metrics for display.
10. The apparatus as in claim 9 , wherein the application experience metric comprises a service level agreement associated with the network path.
11. The apparatus as in claim 9 , wherein the path metrics comprise at least one of: loss, jitter, delay, throughput, or metrics from a particular device associated with the network path.
12. The apparatus as in claim 9 , wherein the apparatus identifies the breaking point in the path metrics by:
causing performance of an experiment along the network path, to obtain one or more samples from the network path, wherein the breaking point is identified based in part on those one or more samples.
13. The apparatus as in claim 12 , wherein the process when executed is further configured to:
recommend the experiment to a user interface, wherein the apparatus causes performance of the experiment in response to an instruction to do so from the user interface.
14. The apparatus as in claim 9 , wherein the process when executed is further configured to:
initiate, based on the breaking point, a reroute of at least a portion of the traffic of the online application.
15. The apparatus as in claim 9 , wherein the application experience metric is predicted by a machine learning model for the online application.
16. A tangible, non-transitory, computer-readable medium storing program instructions that cause a device to execute a process comprising:
obtaining, by the device, path metrics for a network path via which traffic for an online application is conveyed;
modeling, by the device, uncertainty of an application experience metric predicted for the online application based on the path metrics;
identifying, by the device and based on the uncertainty of the application experience metric modeled by the device, a breaking point in the path metrics at which the application experience metric predicted for the online application is expected to switch from being acceptable to being degraded, wherein the breaking point in the path metrics is located in a region of higher uncertainty of the application experience metric than a region where the application experience metric is predicted to be acceptable for the online application; and
providing, by the device, the breaking point in the path metrics for display.