Network trace categorization in content delivery
In some embodiments, a method receives a network trace from a playback session of a delivery of content being managed by a content management system. The network trace is decomposed into a first sub-trace and a second sub-trace based on a threshold. The threshold is used to divide the network trace into the first sub-trace and the second sub-trace. A value is extracted for a feature for the network trace based on the first sub-trace. The method classifies the network trace into a category based on the value and performs an action based on the category for the content management system.
1. A method comprising:
receiving a network trace from a playback session of a delivery of content being managed by a content management system, wherein the network trace captures network conditions during the delivery of the content;
decomposing the network trace into a low frequency sub-trace and a high frequency sub-trace based on a threshold, wherein the threshold is used to divide the network trace into the low frequency sub-trace and the high frequency sub-trace, wherein decomposing the network trace comprises:
converting the network trace into the low frequency sub-trace and the high frequency sub-trace from a time domain to a frequency domain; and
converting the low frequency sub-trace to the time domain;
extracting a value for a feature for the network trace based on the low frequency sub-trace;
classifying the network trace into a category based on the value; and
performing an action based on the category for the content management system.
2. The method of claim 1 , wherein receiving the network trace comprises receiving the network trace from a device that measures values that are included in the network trace during the delivery of the content.
3. The method of claim 1 , wherein decomposing the network trace comprises:
using the threshold to determine the network trace is a low frequency sub-trace and is a high frequency sub-trace in the frequency domain.
4. The method of claim 1 , further comprising:
analyzing the low frequency sub-trace to determine sub-segments based on change points that are detected in the low frequency sub-trace.
5. The method of claim 4 , wherein analyzing the low frequency sub-trace to determine sub-segments comprises:
traversing the low frequency sub-trace to determine changes in the low frequency sub-trace that meet a second threshold.
6. The method of claim 4 , wherein analyzing the low frequency sub-trace to determine sub-segments comprises:
determining a peak of the low frequency sub-trace;
determining a trough of the low frequency sub-trace; and
analyzing the peak and the trough to determine if a second threshold is met to determine a change point occurred.
7. The method of claim 4 , wherein the value is based on one or more values for one or more sub-segments.
8. The method of claim 1 , wherein extracting the value for the feature comprises:
extracting a plurality of values for a plurality of features for trace low frequency sub-trace.
9. The method of claim 8 , wherein classifying the network trace into the category comprises:
using the plurality of values to determine the category.
10. The method of claim 1 , wherein classifying the network trace into the category comprises:
determining a plurality of categories;
determining a level for categories in the plurality of categories; and
using the level for the categories to perform the action.
11. The method of claim 10 , wherein determining the level for categories in the plurality of categories comprises:
selecting a level for each category in the plurality of categories to determine the action.
12. The method of claim 10 , wherein performing the action comprises:
using the level for the plurality of categories to adjust a parameter that is used for delivery of content.
13. The method of claim 12 , wherein the parameter is used for a selection of a bitrate in an adaptive bitrate algorithm or selection of a content delivery network to process a request for content.
14. The method of claim 10 , wherein performing the action comprises:
generating a visualization on a display using the respective level for the plurality of categories.
15. The method of claim 1 , wherein performing the action comprises:
correlating quality of service information for the network trace to the category.
16. A non-transitory computer-readable storage medium having stored thereon computer executable instructions, which when executed by a computing device, cause the computing device to be operable for:
receiving a network trace from a playback session of a delivery of content being managed by a content management system, wherein the network trace captures network conditions during the delivery of the content;
decomposing the network trace into a low frequency sub-trace and a high frequency sub-trace based on a threshold, wherein the threshold is used to divide the network trace into the low frequency sub-trace and the high frequency sub-trace, wherein decomposing the network trace comprises:
converting the network trace into the low frequency sub-trace and the high frequency sub-trace from a time domain to a frequency domain; and
converting the low frequency sub-trace to the time domain;
extracting a value for a feature for the network trace based on the low frequency sub-trace;
classifying the network trace into a category based on the value; and
performing an action based on the category for the content management system.
17. The non-transitory computer-readable storage medium of claim 16 , further operable for:
analyzing the low frequency sub-trace to determine sub-segments based on change points that are detected in the low frequency sub-trace.
18. An apparatus comprising:
one or more computer processors; and
a computer-readable storage medium comprising instructions for controlling the one or more computer processors to be operable for:
receiving a network trace from a playback session of a delivery of content being managed by a content management system, wherein the network trace captures network conditions during the delivery of the content;
decomposing the network trace into a low frequency sub-trace and a high frequency sub-trace based on a threshold, wherein the threshold is used to divide the network trace into the low frequency sub-trace and the high frequency sub-trace, wherein decomposing the network trace comprises:
converting the network trace into the low frequency sub-trace and the high frequency sub-trace from a time domain to a frequency domain; and
converting the low frequency sub-trace to the time domain;
extracting a value for a feature for the network trace based on the low frequency sub-trace;
classifying the network trace into a category based on the value; and
performing an action based on the category for the content management system.