Content adjustment system for reduced latency
AssigneeAmazon Technologies, Inc.
InventorsMin Kyoung Kang, Ronil Sudhir Mokashi
Life of this patent
Estimating the term…
Abstract
Techniques for reducing the latency of content retrieval from a content delivery network include receiving a request from a client device for media content, parsing the request for attributes associated with the request and the client device, and providing the attributes to a machine learning model to perform server-side prediction of an estimated retrieval time of the media content. A quality level for the media content is determined based on the estimated retrieval time, and the requested media content is provided to the client device at the determined quality level.
Claims22
A method performed by a content delivery server, the method comprising:
- obtaining training data by:
- collecting information on historical requests from client devices for streaming video content, and historical content retrieval times for the historical requests;
- and
- extracting a set of attributes from the historical requests;
- training a gradient boosted tree machine learning model using the set of attributes and the historical content retrieval times to perform server-side prediction of estimated retrieval times;
- receiving a new request to stream video content from a client device;
- providing attributes of the new request to the gradient boosted tree machine learning model to predict an estimated retrieval time for the new request;
- mapping the estimated retrieval time for the new request to a resolution for the requested video content;
- and
- streaming the requested video content to the client device at the resolution mapped to the estimated retrieval time.
A method comprising:
- receiving a request from a client device for media content;
- parsing the request for attributes associated with the request and the client device;
- providing the attributes to a machine learning model to perform server-side prediction of an estimated retrieval time of the media content;
- determining a quality level for the media content based on the estimated retrieval time;
- and
- providing the media content to the client device at the determined quality level.
A computing device comprising:
- one or more processors coupled to a memory storing instructions, which when executed by the one or more processors, cause the computing device to perform operations including:
- receiving a request from a client device for media content;
- parsing the request for attributes associated with the request and the client device;
- providing the attributes to a machine learning model to perform server-side prediction of an estimated retrieval time of the media content;
- determining a quality level for the media content based on the estimated retrieval time;
- and
- providing the media content to the client device at the determined quality level.
Dependencies are inferred from the claim text.