Abstract: One embodiment of the present invention sets forth a technique for streaming digital media content with transition effects. Audio data may be modulated and received by the streaming device for an audio fade-in effect. Similarly, video data may be modulated to perform various visual effects. Such a technique allows for fading-in of audio and/or video, fading-out of audio and/or video, and effects that blend audio from different content and video from different content during a transition from one digital media content to a different digital media content or within the same digital media content. The modulated audio and video data to generate the audio or video effect during a transition may be precomputed or computed in real-time and transmitted to the streaming device. The streaming device switches between original unmodulated content streams and effect content streams when transitions are detected.

Abstract: Embodiments of the present invention set forth techniques for a content player to stream a media file using multiple network connections. To stream the media file, the content player downloads metadata associated with a requested media file, establishes a network connection with multiple content servers (or multiple network connections with a single content server or both) and begins requesting portions of the media file. In response, the requested portions are transmitted to the content player. The content player may employ a predictive multi-connection scheduling approach to determine which network connection to use in downloading a given chunk.

Abstract: In one embodiment of the present invention, a native resolution analyzer generates a log-magnitude spectrum that elucidates sampling operations that have been performed on a scene. In operation, the native resolution analyzer performs a transform operation of a color component associated with a frame included in the scene to generate a frame spectrum. The native resolution analyzer then normalizes the magnitudes associated with the frame spectrum and logarithmically scales the normalized magnitudes to create a log-magnitude frame spectrum. This two dimensional log-magnitude frame spectrum serves as a frequency signature for the frame. More specifically, patterns in the log-magnitude spectrum reflect re-sampling operations, such as a down-sampling and subsequent up-sampling, that may have been performed on the frame.

Abstract: In an embodiment, a data store storing a first image and a second image; a computer processor coupled to the data store and programmed to: generate a first fingerprint of the first image based on a first plurality of pixels from the first image stored in the data store; generate a second fingerprint of the second image based on a second plurality of pixels from the second image stored in the data store; determine a closeness value between the first fingerprint and the second fingerprint; determine, based on the closeness value, whether the first image matches the second image.

Abstract: One embodiment of the present invention sets forth a technique for managing playback of digital content on two or more different playback devices. A playback device designated as a controller is associated with a playback device designated as a playback target. Association is implemented via a local network discovery protocol. The controller issues commands to the target, which then streams digital content directly from a content server. The controller is able to separately stream related digital content for independent, but synchronized playback. Server mediated operations, such as device activation and application installation, are advantageously enabled by explicit association between the controller and the target.

Abstract: One embodiment of the present invention sets forth a technique for transmitting a digital content stream through a network between at least one content server and a content player. The method includes communicating with the at least one content server to set up a plurality of parallel network connections on the network to transmit the digital content stream. The method further includes sending a first data request and a second data request within a first time limit via at least one tested network connection in the plurality of parallel network connections to evaluate whether pipelining of the digital content stream is possible for the at least one tested network connection. The method also includes utilizing pipelining to transmit the digital content stream over the at least one tested network connection based on a first response to the first data request and a second response to the second data request.

Abstract: A data processing method, comprising: using computing apparatus, causing instantiating a plurality of baseline application instances that are running a first version of software, and one or more canary application instances that are running a second version of the software; using computing apparatus including a load balancer, causing selectively routing a first proportion of incoming requests to the baseline instances, and routing a second proportion of the incoming requests to the canary instances; monitoring the plurality of canary instances to collect performance data for performance metrics; determining that the performance data indicates a negative performance issue, and in response thereto: using computing apparatus, automatically updating the first proportion to be larger and updating the second proportion to be smaller, and then reconfiguring the load balancer based upon the first proportion and the second proportion; terminating one or more of the canary application instances.

Abstract: Techniques are disclosed for generating preference rankings in response to requests for streaming media content received from client devices. The preference rankings are used to indirectly partition streaming media content across different content distribution networks (CDNs). Such partitions may be referred to as a “soft” partition, as all of the streaming media content remains available from each CDN. That is, the partitioning of content across CDNs is not absolute, and any content item may be requested from any CDN. Doing so ensures that content remains available even where the preferred CDN for a given content item fails, while at the same time increasing the probability of a cache hit at the CDNs, without increasing the cost or the actual cache sizes of the content caches of different CDNs.

Abstract: The disclosure describes a proxy server application that supports the dynamic modification of proxy rules implemented by a proxy server. The proxy rules implemented by the proxy server specify network behaviors to be performed at various points during the handling of requests from client applications. A proxy server implements the proxy rules by processing one or more user-generated network traffic filters for managing network traffic. In an embodiment, users generate network traffic filters by creating network traffic filter source code that specify processing steps to be performed by a proxy server relative to network messages the proxy server receives. In an embodiment, user-generated network traffic filters may be added, removed, reordered, or otherwise modified in a proxy server application at runtime in order to respond to current network conditions or to achieve other desired proxy configurations.

Abstract: Techniques are described for detecting faults in media content based on the behavior of users viewing the media content. Embodiments stream a first instance of media content to one or more users. The behavior of the one or more users is monitored while the users are viewing the streaming first instance of media content. Embodiments then determine whether the first instance of media content is faulty, based on the monitored behavior of the one or more users.

Abstract: One embodiment of the present invention sets forth a technique for providing a variation of digital content that is based on a client-specified context token. Localized digital assets, e.g., graphics resources, video resource, audio resources, and text string resources, are stored as a collection of records and are dynamically linked to a assemble a software system based on the client-specified context token to generate the variation of digital content. The client receives the assembled variation of the digital content including the localized digital assets specified by the client-provided context.

Abstract: One embodiment of the present invention sets forth a technique for adapting playback bit rate to available delivery bandwidth in a content delivery system comprising a content server and a content player. A content player periodically estimates whether a given playback bit rate can feasibly provide complete playback for a given title assuming currently available bandwidth. If playback becomes unfeasible at a current bit rate assuming currently available bandwidth, then the content player adapts the bit rate downward until a feasible bit rate is achieved. If playback is feasible using a higher bit rate, then the content player may adapt the bit rate upward.

Abstract: One embodiment of the present invention sets forth a technique for deferring license retrieval when streaming digital media content. The perceived delay between the time a user selects the protected digital media content to when playback of the protected digital media content begins is reduced because retrieval and playback of an unprotected version of a portion of the digital media content starts before the license and protected version of the digital media content is received. The unprotected version includes fast start streams of audio and video data that may encoded at a lower bit rate than the protected version in order to quickly transfer the fast start streams from the content server to the playback device.

Abstract: Approaches, techniques, and mechanisms are disclosed for implementing a distributed firewall. In an embodiment, many different computer assets police incoming messages based on local policy data. This local policy data is synchronized with global policy data. The global policy data is generated by one or more separate analyzers. Each analyzer has access to message logs, or information derived therefrom, for groups of computer assets, and is thus able to generate policies based on intelligence from an entire group as opposed to an isolated asset. Among other effects, some of the approaches, techniques, and mechanisms may be effective even in computing environments with limited supervision over the attack surface, and/or computing environments in which assets may need to make independent decisions with respect to how incoming messages should be handled, on account of latency and/or unreliability in connections to other system components.

Abstract: Embodiments provide techniques generating and managing encryption keys within a computing infrastructure. Embodiments provide a key publisher that generates and maintains key pairs in a list at a configurable interval. In addition, the key publisher publishes the list to other components within the computing infrastructure. Embodiments also provide a key consumer that downloads the list of encrypted key pairs and maintains an active window of keys to can be accepted from client devices that communicate sensitive data to the computing infrastructure. If the key consumer receives a key from a client device that is outside of the active window yet that corresponds to a future key pair in the list, the key consumer advances the active window towards the future key pair.

Abstract: A method and system for building a point-in-time snapshot of an eventually-consistent data store. The data store includes key-value pairs stored on a plurality of storage nodes. In one embodiment, the data store is implemented as an Apache® Cassandra database running in the “cloud.” The data store includes a journaling mechanism that stores journals (i.e., inconsistent snapshots) of the data store on each node at various intervals. In Cassandra, these snapshots are sorted string tables that may be copied to a back-up storage location. A cluster of processing nodes may retrieve and resolve the inconsistent snapshots to generate a point-in-time snapshot of the data store corresponding to a lagging consistency point. In addition, the point-in-time snapshot may be updated as any new inconsistent snapshots are generated by the data store such that the lagging consistency point associated with the updated point-in-time snapshot is more recent.

Abstract: Techniques are described for identifying a root cause of a pattern of performance data in a system including a plurality of services. Embodiments provide dependency information for each of the plurality of services, where at least one of the plurality of services is dependent upon a first one of the plurality of services. Each of the plurality of services is monitored to collect performance data for the respective service. Embodiments further analyze the performance data to identify a cluster of services that each follow a pattern of performance data. The first one of the services in the cluster of services is determined to be a root cause of the pattern of performance data, based on the determined dependency information for each of the plurality of services.

Abstract: Techniques for adaptive metric collection, metric storage, and alert thresholds are described. In an approach, a metric collector computer processes metrics as a collection of key/value pairs. The key/value pairs represent the dimensionality of the metrics and allows for semantic queries on the metrics based on keys. In an approach, a storage controller computer maintains a storage system with multiple storage tiers ranked by speed of access. The storage computer stores policy data that specifies the rules by which metric records are stored across the multiple storage tiers. Periodically, the storage computer moves database records to higher or lower tiers based on the policy data. In an approach, a metric collector in response to receiving a new metric, generates a predicted metric value based on previously recorded metric values and measures the deviation from the new metric value to determine whether an alert is appropriate.

Abstract: Techniques are disclosed for reclaiming resources within a distributed computing system. A reclamation application searches the distributed computing architecture for unused resources, classifies the unused resources, and determines an expiration period based on the classification. The reclamation application determines a candidate owner of the resource based on one or more characteristics of the resource. The reclamation application then notifies the candidate owner that the resource is to be reclaimed unless claimed by the candidate owner within the expiration period. If the candidate owner claims the resource within the expiration period, then the reclamation application terminates the reclamation of the resource. If the candidate owner does not claim the resource within the expiration period, then the reclamation application reclaims the resource after the duration of the expiration period.

Abstract: In an approach, a server computer receives a request from a client computer specifying particular content for a particular user, wherein the particular content is associated with an original audio language. In response to receiving the request, the server computer selects a preferred audio language and a preferred subtitle language for the particular content based on a particular record of a preference database. The server computer returns asset identifying data that the client computer uses to obtain a stream of the particular content using the preferred audio language and the preferred subtitle language from a content delivery network (CDN) or other asset location. The server computer receives a message from the client computer that identifies an presented audio language and a presented subtitle language that were presented to the particular user while the particular content streamed.