Abstract: In one embodiment of the present invention, an encode validator identifies and classifies errors introduced during the parallel chunk-based translation of a source to a corresponding aggregate encode. In operation, upon receiving a source for encoding, a frame difference generator creates a frame difference file for the source. A parallel encoder then distributes per-chunk encoding operations across machines and creates an aggregate encode. The encode validator decodes the aggregate encode and creates a corresponding frame difference file. Subsequently, the encode validator performs phase correlation operations between the two frame difference files to detect errors generated by encoding process faults (i.e., dropping a frame, etc.) while suppressing discrepancies inherent in encoding, such as those attributable to low bit-rate encoding.

Abstract: One embodiment of the invention disclosed herein provides techniques for distributing tasks to skilled workers participating in a managed crowd-sourcing workforce. A crowd worker routing system receives a request for a task assignment from a first worker participating in the crowd-sourcing workforce. The crowd worker routing system retrieves one or more scores associated with the first worker, wherein each score is related to the performance of the first worker with respect to completing a prior set of task assignments. The crowd worker routing system determines that a first task assignment included in a first set of task assignments is related to a first score included in the one or more scores. The crowd worker routing system determines whether to route the first task assignment to the first worker for completion based on the first score and one or more routing rules.

Abstract: An encoding engine encodes a video sequence to provide optimal quality for a given bitrate. The encoding engine cuts the video sequence into a collection of shot sequences. Each shot sequence includes video frames captured from a particular capture point. The encoding engine resamples each shot sequence across a range of different resolutions, encodes each resampled sequence with a range of quality parameters, and then upsamples each encoded sequence to the original resolution of the video sequence. For each upsampled sequence, the encoding engine computes a quality metric and generates a data point that includes the quality metric and the resample resolution. The encoding engine collects all such data points and then computes the convex hull of the resultant data set. Based on all convex hulls across all shot sequences, the encoding engine determines an optimal collection of shot sequences for a range of bitrates.

Abstract: Increasingly, children are consuming digital content on computers, laptops, tablets, and even smartphones. The sheer volume of content available to children can be daunting for parents to manage and filter for their children. Primitive filtering mechanisms based on a target audience age can be used to eliminate some content based on a child's chronological age. However, children develop at different rates so that even children of the same chronological age can have different developmental stages and so have different preferences for content they consume. Systems and methods described herein provide children with access to developmental stage-appropriate content in a developmental stage-appropriate user interface. The developmental stage-appropriate interface includes representations of the content that are developmental stage-appropriate and surrounding elements that are also developmental stage-appropriate.

Abstract: A system for managing memory input/output management is provided herein. The system may include a processor and a memory storing machine-readable instructions that when executed by the processor, cause the system to perform operations including receiving read requests for first data stored on a solid-state storage drive, receiving write requests for second data to be stored on the storage drive, placing at least some of the read requests in a first queue, placing at least some of the write requests in a second queue, the second queue having a size limit, processing data at a disk driver layer from the first queue and the second queue in a manner such that selection of a request from either queue is biased towards the first queue. Associated methods are also included.

Abstract: One embodiment of the present invention sets forth a technique for generating one or more hash data structures. The technique includes generating a hash data structure having entries that correspond to a plurality of content servers, and, for each file included in a first plurality of files, allocating the file to one or more content servers included in the plurality of content servers by comparing a hash value associated with the file to one or more entries included in the entries. The technique further includes comparing a network bandwidth utilization of a first content server to a network bandwidth utilization associated with one or more other content servers included in the plurality of content servers to generate a result, and modifying a first number of entries associated with the first content server and included in the entries based on the result to generate a biased hash data structure.

Abstract: In various embodiments, a subjective modeling engine mitigates inaccuracies in subjective content assessments. The subjective modeling engine generates a model that includes the subjective content assessments in addition to parameters for subjective scores and subjective factors. The subjective modeling engine initializes the parameters and then performs optimization operations that increase the likelihood that the optimized subjective scores compensate for the optimized subjective factors. Advantageously, because the subjective modeling engine jointly optimizes the subjective scores and the subjective factors, the optimized subjective scores provide unbiased and consistent digital content assessments.

Abstract: The disclosed apparatus may include a rack-side support structure dimensioned to hold a two-sided port interface with a rack-side mating end and an adapter-side mating end. The rack-side mating end may be configured to interface with supply cables, and the adapter-side mating end may be configured to interface with an opposite adapter-side mating end of another port interface. The apparatus may also include a device-side support structure dimensioned to hold a two-sided port interface including an opposing adapter-side mating end and a device-side mating end. The opposing adapter-side mating end may be configured to interface with the adapter-side mating end of the rack-side's port interface, and the device-side mating end may interface with cables that connect to the electronic devices. The rack-side support structure may be configured to interlock with the device-side support structure to connect to the electronic devices. Various other methods, systems, and computer-readable media are also disclosed.

Abstract: A method includes receiving, with a computing system, a video item. The method further includes identifying a first set of features within a first frame of the video item. The method further includes identifying, with the computing system, a second set of features within a second frame of the video item, the second frame being subsequent to the first frame. The method further includes determining, with the computing system, differences between the first set of features and the second set of features. The method further includes assigning a clip category to a clip extending between the first frame and the second frame based on the differences.

Abstract: One embodiment of the present invention sets forth a technique for incorporating a text-containing image into a digital. The technique includes analyzing a digital image to determine one or more text-allowed regions included in the digital image, and, for each of the one or more text-allowed regions, computing a numeric value based on a color contrast between pixels of a text-containing image and pixels of the text-allowed region, wherein the text-containing image is to be incorporated into one of the text-allowed regions included in the digital image. The technique further includes selecting a first text-allowed region based at least in part on the numeric value computed for each text-allowed region, and incorporating the text-containing image into the first text-allowed region included in the digital image.

Abstract: A sequence analyzer compares different episodes of an episodic serial to identify portions of a current episode of the serial that have already been played back to a user. Those portions may include introductory material such as credits, or a recap section that includes content from previous episodes. The sequence analyzer parses previous episodes of the serial and selects a representative frame for each shot sequence. The sequence analyzer then generates a fingerprint for each shot sequence based on the associated representative frame. The sequence analyzer compares fingerprints associated with a current episode of the serial to fingerprints associated with one or more previous episodes of the serial to identify shot sequences that have already been played. The user may then skip those repeated sequences via a playback interface.

Abstract: A computer-implemented method and a system for implementation the method are provided. The method includes receiving user activity data for a first set of users of a streaming media system, in which the first set of users having access to a first catalog. Additional user activity data for a second set of users is received. The second set of users has access to a second catalog, which includes media items also present in the first catalog. The second catalog further includes an exclusive media item that is not included in the first catalog. A probability that a user of the first set of users would play the exclusive media item if presented by the streaming media system with the exclusive media item is determined. Information is provided to a user of the second set of users as a recommendation to play the exclusive media item based on the probability.

Abstract: Techniques for predictively scaling a distributed application are described. Embodiments could monitor performance of an application within a cloud computing environment over a first window of time to collect historical performance data. Here, the application comprises a plurality of application instances. A workload of the application could be monitored over a second window of time to collect historical workload data. Embodiments could analyze both the historical performance data and the historical workload data to determine one or more scaling patterns for the application. Upon determining a present state of the application matches one of the one or more scaling patterns, a plan for predictively scaling the application could be determined. Embodiments could then predictively scale the plurality of application instances, based on the determined plan.

Abstract: A bitrate allocation engine allocates bitrates for distributed encoding of source data. Upon receiving a chunk of source data, the bitrate allocation engine generates a curve based on multiple points that each specify a different visual quality level and corresponding encoding bitrate for encoding the chunk. Subsequently, the bitrate allocation engine computes an optimized encoding bitrate based on the generated curve and an optimization factor that is associated with different visual quality levels and corresponding encoding bitrates for multiple chunks of the source data. The bitrate allocation engine then causes the chunk to be encoded at the optimized encoding bitrate. Advantageously, the resulting encoded chunk is optimized with respect to the optimization factor for multiple chunks of the source data.

Abstract: One embodiment of the present invention sets forth a method for updating content stored in a cache residing at an internet service provider (ISP) location that includes receiving popularity data associated with a first plurality of content assets, where the popularity data indicate the popularity of each content asset in the first plurality of content assets across a user base that spans multiple geographic regions, generating a manifest that includes a second plurality of content assets based on the popularity data and a geographic location associated with the cache, where each content asset included in the manifest is determined to be popular among users proximate to the geographic location or users with preferences similar to users proximate to the geographic location, and transmitting the manifest to the cache, where the cache is configured to update one or more content assets stored in the cache based on the manifest.

Abstract: Techniques for transmitting a digital content stream through a network between a plurality of content servers and a content player. Network performance information for the plurality of content servers is retrieved. Embodiments then determine a mapping of the plurality of content servers to a plurality of portions of a content buffer on the content player, based on the retrieved network performance information. A request is transmitted to each of the plurality of content servers to transmit digital content associated with the corresponding portion of the content buffer.

Abstract: A data processing method comprising obtaining a plurality of computer network security threat feeds from two or more computer threat detection systems; based upon computer network attack information in the computer network security threat feeds, determining a threat score that represents a severity of an actual or suspected attack on a particular host in a computer network; obtaining an asset value for the particular host that indicates a worth of the particular host, and updating the threat score based upon the asset value; mapping the updated threat score to one of a plurality of remediation actions, wherein a first remediation action is mapped when the updated threat score is low and a second, different remediation action is mapped when the updated threat score is high; based upon the updated threat score and the mapping, selecting and automatically performing one of the plurality of remediation actions on the particular host; wherein the method is performed by one or more special-purpose computing devices

Abstract: A certificate orchestration system for digital certificate and encryption key management is provided herein along with associated methods. The system includes a certificate orchestration server having a processing device in communication with a coupled storage system that is coupled to the certificate orchestration server. The system further includes an interface provided by the certificate orchestration server to a client device; and a database to store digital certificates and keys. The certificate orchestration server is configured to receive a request from the client device to generate a public key, receive the public key from a third-party certificate authority system over an external network, store the public key in the coupled storage system. The coupled storage system is not directly connected to the client device.

Abstract: During a Transmission Control Protocol (“TCP”) session, a sending endpoint computer monitors amounts of data sent and patterns of data loss as data is sent to a receiving endpoint computer. The sending endpoint computer periodically determines whether data is being sent below, at or above path capacity, based on the monitored amounts of data sent and patterns of data loss observed. The sending endpoint computer periodically dynamically adjusts the rate at which data is sent to the receiving endpoint computer, in response to the determinations whether data is being sent below, at or above path capacity.

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.