Dynamic mechanism for migrating traffic spikes in a streaming media network

1 . A system for dynamically migrating traffic spikes in a streaming media network, the system comprising:

a computing platform including a processing hardware and a system memory;

the system memory storing a software code;

a machine-learning (ML) model having a deep neural network (NN) trained, using content consumption histories and demographic data associated with a plurality of user accounts, to identify one or more of the plurality of user accounts likely to request content upon release;

the processing hardware configured to execute the software code to:

receive content data identifying a new content and a future release date for the new content;

provide a prediction, using the ML model, based on a weighted combination of a plurality of previous actions by respective users of the plurality of user accounts, wherein one of the plurality of previous actions included in the weighted combination is a previous attempt to access or preview another content before release of the another content, the prediction identifying a subset of the plurality of user accounts likely to request the new content on or after the future release date;

prioritize, based on the prediction, a schedule for pre-provisioning of respective authentication data for accessing the new content to each user account of the subset of the plurality of user accounts; and

pre-provision, using the prioritized schedule, the respective authentication data to a user system of a user corresponding to each user account of the subset of the plurality of user accounts by transmitting the respective authentication data to the user system prior to the future release date of the new content, the respective authentication data including authentication credentials for a respective one user account of the subset of the plurality of user accounts and playback data for the new content, the playback data including one or more playlists for the new content, the authentication credentials and the playback data being encrypted and constrained by a usage rule preventing decryption prior to the future release data;

wherein pre-provisioning of the respective authentication data reduces a processing overhead of a streaming platform providing the new content, on the future release date, by enabling the user system to authenticate and obtain the playback data locally without contacting the streaming platform upon release of the new content, and wherein a performance of the system to dynamically migrate the traffic spikes in the streaming media network is improved by adjusting one or more weights of the weighted combination used to provide the prediction.

2 . The system of claim 1 , wherein the respective authentication data enables each user account of the subset of the plurality of user accounts to receive the new content concurrently.

3 . The system of claim 1 , wherein the processing hardware is further configured to execute the software code to:

cluster, based on respective user histories of each of the plurality of user accounts, another subset of the plurality of user accounts into a user group; and

provide, using the trained ML model, the prediction identifying the subset of the plurality of user accounts likely to request the new content upon release of the new content further based on the user group.

4 . The system of claim 1 , wherein the processing hardware is further configured to execute the software code to:

prioritize the schedule for pre-provisioning of the respective authentication data to each user account of the subset of the plurality of user accounts further based on respective user histories of each user account of the subset of the plurality of user accounts.

5 . The system of claim 1 , wherein the processing hardware is further configured to execute the software code to:

prioritize the schedule for pre-provisioning of the respective authentication data to each user account of the subset of the plurality of user accounts further based on respective geo-locations of each user account of the subset of the plurality of user accounts.

6 . The system of claim 1 , wherein the respective authentication data includes a decryption key.

7 . The system of claim 1 , wherein the processing hardware is further configured to execute the software code to:

provide, using the prioritized schedule, the new content to each user account of the subset of the plurality of user accounts prior to release of the new content to other user accounts.

8 . The system of claim 1 , wherein the processing hardware is further configured to execute the software code to:

receive, before the future release date of the new content, a plurality of content access requests for the new content from another plurality of user accounts;

partition at least some of the other plurality of user accounts into a first distribution group and a second distribution group;

enable user accounts included in the first distribution group to receive the new content during a predetermined time interval triggered by the future release date of the new content; and

prevent user accounts included in the second distribution group from receiving the new content until the predetermined time interval elapses.

9 . The system of claim 1 , wherein the new content comprises streaming media content including at least one of a movie, television (TV) programming content, or a video game.

10 . A method for use by system including a machine-learning (ML) model having a deep neural network (NN) trained, using content consumption histories and demographic data associated with a plurality of user accounts, to identify one or more of the plurality of user accounts likely to request content upon release, to dynamically migrate traffic spikes in a streaming media network, the method comprising:

receiving content data identifying a new content and a future release date for the new content;

generating, using the ML model, based on a weighted combination of a plurality of previous actions by respective users of the plurality of user accounts, wherein one of the plurality of previous actions included in the weighted combination is a previous attempt to access or preview another new content before release of the another new content, a prediction identifying a subset of the plurality of user accounts likely to request the new content on or after the future release date;

prioritizing a schedule for pre-provisioning of respective authentication data for accessing the new content to each user account of the subset of the plurality of user accounts; and

pre-provisioning, based on prioritizing, the respective authentication data to a user system of a user corresponding to each user account of the subset of the plurality of user accounts by transmitting the respective authentication data to the user system prior to the future release date of the new content, the respective authentication data including authentication credentials for a respective one user account of the subset of the plurality of user accounts and playback data for the new content, the playback data including one or more playlists for the new content, the authentication credentials and the playback data being encrypted and constrained by a usage rule preventing decryption prior to the future release data;

wherein pre-provisioning of the respective authentication data reduces a processing overhead of a streaming platform providing the new content, on the future release date, by enabling the user system to authenticate and obtain the playback data locally without contacting the streaming platform upon release of the new content, and wherein a performance of the system to dynamically migrate the traffic spikes in the streaming media network is improved by adjusting one or more weights of the weighted combination used to provide the prediction.

11 . The method of claim 10 , wherein the respective authentication data enables each user account of the subset of the plurality of user accounts to receive the new content concurrently.

12 . The method of claim 10 , further comprising:

clustering, based on respective user histories of each of the plurality of user accounts, another subset of the plurality of user accounts into a user group; and

providing, using the trained ML model, the prediction identifying the subset of the plurality of user accounts likely to request the new content upon release of the new content further based on the user group.

13 . The method of claim 10 , further comprising:

prioritizing the schedule for pre-provisioning of the respective authentication data to each user account of the subset of the plurality of user accounts further based on respective user histories of each user of the subset of the plurality of user accounts.

14 . The method of claim 10 , further comprising:

prioritizing the schedule for pre-provisioning of the respective authentication data to each user account of the subset of the plurality of user accounts further based on respective geo-locations of each user account of the subset of the plurality of user accounts.

15 . The method of claim 10 , wherein the respective authentication data includes a decryption key.

16 . The method of claim 10 , further comprising:

providing, using the prioritized schedule, the new content to each user account of the subset of the plurality of user accounts prior to the release of the new content to other user accounts.

17 . The method of claim 10 , further comprising:

receiving, before the future release date of the new content, a plurality of content access requests for the new content from another plurality of user accounts;

partitioning at least some of the other plurality of user accounts into a first distribution group and a second distribution group;

enabling user accounts included in the first distribution group to receive the new content during a predetermined time interval triggered by the future release date of the new content; and

preventing user accounts included in the second distribution group from receiving the new content until the predetermined time interval elapses.

18 . The method of claim 10 , wherein the new content comprises streaming media content including at least one of a movie, television (TV) programming content, or a video game.