Using throughput mode distribution as a proxy for quality of experience and path selection in the internet
In one embodiment, a device calculates one or more distributions of bitrates associated with an application whose traffic is conveyed via one or more paths in a network. The device detects throughput modes of the application, based on the one or more distributions of bitrates associated with the application. The device associates each throughput mode with a quality of experience label, to form a plurality of pairs of throughput modes and quality of experience labels. The device estimates a quality of experience metric for the application, based on a bitrate of the application and the plurality of pairs of throughput modes and quality of experience labels.
1. A method comprising:
calculating, by a device, one or more distributions of bitrates associated with an application whose traffic is conveyed via one or more paths in a network;
detecting, by the device, throughput modes of the application, based on the one or more distributions of bitrates associated with the application;
associating, by the device, each throughput mode with a quality of experience label, to form a plurality of pairs of throughput modes and quality of experience labels; and
estimating, by the device, a quality of experience metric for the application, based on a bitrate of the application and the plurality of pairs of throughput modes and quality of experience labels.
2. The method as in claim 1 , wherein the network comprises a software-defined wide area network (SD-WAN) and wherein the one or more paths comprise one or more network tunnels.
3. The method as in claim 1 , wherein the quality of experience labels indicate whether the quality of experience of the application is acceptable or unacceptable.
4. The method as in claim 1 , wherein the device calculates the one or more distributions of bitrates on a per-flow and per-destination basis.
5. The method as in claim 1 , wherein associating each throughput mode with a quality of experience label, to form the plurality of pairs of throughput modes and quality of experience labels, comprises:
estimating the throughput modes from the one or more distributions, based on a predefined number of throughput modes for the application.
6. The method as in claim 1 , wherein associating each throughput mode with a quality of experience label, to form the plurality of pairs of throughput modes and quality of experience labels, comprises:
identifying local maxima of the one or more distributions as the throughput modes.
7. The method as in claim 1 , wherein each of the one or more distributions of bitrates is associated with a range of measured loss.
8. The method as in claim 1 , further comprising:
initiating, by the device and based on the quality of experience metric estimated for the application, a rerouting of traffic of the application from the one or more paths in the network to one or more other paths.
9. The method as in claim 1 , further comprising:
retrieving a quality of experience measurement from the application; and
comparing the quality of experience measurement from the application to the quality of experience metric estimated for the application.
10. The method as in claim 1 , further comprising:
using the quality of experience metric estimated for the application to train a forecasting engine configured to forecast the quality of experience metric and proactively reroute traffic of the application, based on that forecast.
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:
calculate one or more distributions of bitrates associated with an application whose traffic is conveyed via one or more paths in a network;
detect throughput modes of the application, based on the one or more distributions of bitrates associated with the application;
associate each throughput mode with a quality of experience label, to form a plurality of pairs of throughput modes and quality of experience labels; and
estimate a quality of experience metric for the application, based on a bitrate of the application and the plurality of pairs of throughput modes and quality of experience labels.
12. The apparatus as in claim 11 , wherein the network comprises a software-defined wide area network (SD-WAN) and wherein the one or more paths comprise one or more network tunnels.
13. The apparatus as in claim 11 , wherein the quality of experience labels indicate whether the quality of experience of the application is acceptable or unacceptable.
14. The apparatus as in claim 11 , wherein the apparatus calculates the one or more distributions of bitrates on a per-flow and per-destination basis.
15. The apparatus as in claim 11 , wherein the apparatus associates each throughput mode with a quality of experience label, to form the plurality of pairs of throughput modes and quality of experience labels, by:
estimating the throughput modes from the one or more distributions, based on a predefined number of throughput modes for the application.
16. The apparatus as in claim 11 , wherein the apparatus associates each throughput mode with a quality of experience label, to form the plurality of pairs of throughput modes and quality of experience labels, by:
identifying local maxima of the one or more distributions as the throughput modes.
17. The apparatus as in claim 11 , wherein the process when executed is further configured to:
initiate, based on the quality of experience metric estimated for the application, a rerouting of traffic of the application from the one or more paths in the network to one or more other paths.
18. The apparatus as in claim 11 , wherein the process when executed is further configured to:
retrieving a quality of experience measurement from the application; and
comparing the quality of experience measurement from the application to the quality of experience metric estimated for the application.
19. The apparatus as in claim 11 , wherein the process when executed is further configured to:
use the quality of experience metric estimated for the application to train a forecasting engine configured to forecast the quality of experience metric and proactively reroute traffic of the application, based on that forecast.
20. A tangible, non-transitory, computer-readable medium storing program instructions that cause a device to execute a process comprising:
calculating, by the device, one or more distributions of bitrates associated with an application whose traffic is conveyed via one or more paths in a network;
detecting, by the device, throughput modes of the application, based on the one or more distributions of bitrates associated with the application;
associating, by the device, each throughput mode with a quality of experience label, to form a plurality of pairs of throughput modes and quality of experience labels; and
estimating, by the device, a quality of experience metric for the application, based on a bitrate of the application and the plurality of pairs of throughput modes and quality of experience labels.