Amazon Technologies, Inc.

Ayush Goyal

Arvinth Ravi

Carl Wesley Summers

Zongyi Liu

Bruce Ferry

Kun Chen

Nataraj Mocherla

Jungwoo Jang

Ravishankar Rajagopal

Adam Winstanley

Systems, methods, and computer-readable media are disclosed for systems and methods for latency detection for streaming video. Example methods may include receiving a request to determine a time to first frame for video content, determining a streaming service and a playback device associated with the request, and determining a time at which playback of the content is initiated. Example methods include capturing a playback sequence of the video content, classifying individual frames of the playback sequence, where each classification is associated with a probability value indicative of a likelihood the frame is a first frame, analyzing classifications of the individual frames, and determining, based at least in part on the analysis of the classifications of the individual frames, the time to first frame for the video content.

Latency detection for streaming video

A context management service provides historical context information describing the state of a computer system at a particular time to an event processing service. When an event is received by the event processing service, the event processing service determines, based at least in part on the event, a point in time for which context information is desired. The event processing service requests the desired context information from the context management service. The context management service produces a context snapshot that describes the state of the computer system at the determined point in time, and provides the context snapshot to the event processing service. The event processing service is able to use the context information to process the event in accordance with the state of the computer system at the determined point in time.

Event context service

Computing resource service providers provide computing resources to a plurality of customers. To avoid customers experiencing a degradation in service associated with the computing resources provided to customers, a main queue and a sideline queue may be used to manage and distribute customer events to service endpoints. Customer events may be placed in a main queue and transmitted, by a delivery host, to a service endpoint. If the delivery host receives a throttle response from the service endpoint, the delivery host may enqueue the customer event in a sideline queue and generate and/or store state information associated with the customer event. The state information may include an interval of time at the expiration of which the customer event may be retransmitted to the service endpoint.

Multiple queueing for distributed environments

Growing community of inventors

Seattle, WA, United States of America

Ayush Goyal