Abstract: A distributed system may comprise a plurality of computing nodes, each of which may provide computing capacity for operating various computing services. The system may maintain a memory-based store of records representative of capacity available on the computing nodes. A record indicative of a computing node having a requested amount of capacity may be selected. The record may be modified to indicate that the computing node is a candidate for providing the requested amount of capacity. Upon receiving information confirming that the computing node has been configured to provide the requested capacity, the record may be modified to indicate that the computing node has committed to provide the requested capacity.

Abstract: An image analysis system automatically determines a positioning and orientation of objects in an image. The image analysis system may be configured to evaluate images and determine whether the orientation and position of objects depicted in the images meet acceptable parameters. The image analysis system may process image data associated with an image to identify edges present in the image, and determine object lines based on the edges. Once the lines are identified, the image analysis system determines vanishing points for sets of the object lines. The image analysis system may then estimate camera parameters and determine the orientation of the objects as depicted in the image based on the vanishing points.

Abstract: Disclosed are embodiments for creating and provisioning instances having an expiration time. One exemplary embodiment comprises receiving a request to launch at least one instance, the request associated with an expiration time. In response to the request, an an instance can be instantiated using one or more server computers. Further, the embodiment includes determining, based at least in part on the expiration time, to terminate the provisioned instance and subsequently terminating the provisioned instance.

Abstract: A control node of a multi-tenant stream processing service receives a request indicating an operation to be performed on data records of a particular data stream. Based on a stream partitioning policy, the control node determines an initial number of worker nodes to be used. The control node configures a worker node to perform the operation on received records. In response to a determination that the worker node is in an unhealthy state, the control node configures a replacement worker node.

Abstract: Systems and methods for generating a programmatic implementation based on a set of recorded API calls. One example includes recording a set of user actions performed via a temporary second user account and generating a programmatic implementation based at least in part on the recorded user actions.

Abstract: A system in which a device may automatically provision another device with credentials, at the behest of a cloud-based service, based in part on the physical proximity of the device to be provisioned. The provisioning device and the device to be provisioned may use a radio access technology (RAT) with a limited radio range. Account information associated with the device to be provisioned is known to the cloud-based service, which authenticates the device to be provisioned via the device with credentials.

Abstract: Systems and methods are described for management of data transmitted between computing devices in a communication network. An administrative component can configure one or more devices in the communication path of messages to be exchanged by devices to interpret codes embedded in the communication messages. A receiving device can review incoming messages and process the message to determine how the communication message will be routed or otherwise processed according to business rules or logic.

Abstract: Techniques are described for handling temporal overlaps that occur between fragments of streaming media content, such as between a fragment of primary media content (e.g., a live sporting event) and a fragment of secondary media content (e.g., ad media content) dynamically inserted into the stream.

Abstract: A technology is provided for delivering a data object to device. Data delivery instructions to distribute a data object to a device may be received at a delivery service and include a storage location of the data object, a messaging topic, and a size of data receivable by the device. The data object is then divided into blocks corresponding to the size of data receivable by the device. A messaging topic is identified to which the device is subscribed via a messaging service. The blocks are sent to the device using the messaging topic and a network protocol installed on the device.

Abstract: A game spectating system that provides interfaces and methods for providing game content to viewers. The spectating system obtains game metadata from game systems for games being broadcast by the spectating system according to the API, and generates broadcast content based at least in part on the game metadata. The broadcast content includes user interface (UI) elements for obtaining virtual or physical objects corresponding to the game content. The broadcast content is provided to spectator devices with the broadcasts. The spectating system receives indications of spectators' interactions with the broadcast content from the spectator devices including interactions requesting the virtual or physical objects, and facilitates provisioning or delivery of the objects to respective spectator devices or spectators.

Abstract: Methods, systems, and computer-readable media for request authorization using service coordination are disclosed. An authorization data structure and an operation data structure are selected based at least in part on a request for an operation. The authorization data structure comprises a directed acyclic graph representing a flow of data between service operations associated with authorization of the operation, and the operation data structure comprises a directed acyclic graph representing a flow of data between a service operations associated with execution of the operation. Authorization of the operation is attempted using the authorization data structure, comprising invoking one or more of the service operations associated with authorization. If the authorization is successful, then the execution of the operation is initiated using the operation data structure, comprising invoking one or more of the service operations associated with execution.

Abstract: An audio/video (A/V) recording and communication doorbell device includes an input port, a switch, a first power supply, a second power supply, a button, a first controller, and a second controller. The switch is electrically coupled across the input port. The first power supply receives power from the input port and powers a first power supply rail, and the second power supply powers a second power supply rail. The button, when pressed, activates a signaling device. The first controller is at least partially powered from the first power supply rail and closes the switch in response to the button being pressed. The second controller is at least partially powered from the second power supply rail.

Abstract: Techniques described and suggested include systems, methods, and processes for determining data-related attributes, such as size, for data and/or data containers (such as volumes and file systems) resident on distributed storage systems. For example, a shared records storage device may be utilized to gather information related to file systems stored on a distributed storage system having a plurality of storage servers. A data processor may be used to generate information records and meter data storage size for file systems resident on the distributed storage system, and such records may be provided to requestors, such as customers associated with the file systems.

Abstract: The disclosure is related to allocation of bits in a media stream. In an example, a video stream is segmented into groups of pixels. A determination of a class type is made for individual ones of the groups of pixels. The determination can be based at least in part on semantic similarity of the class type and of a scene represented in the groups of pixels. A further determination occurs for sets of classified data associated with regions of interest (ROIs) according to the determined class type. Masking data associated with the sets of classified data is provided and confidence measures associated with the sets of classified data and the ROIs are determined. Bits are then allocated for groups of pixels based on the masking data and the confidence measures. Thereafter, a bit stream with the bits can be transmitted for playback on a computing device.

Abstract: Techniques for determining the state of data in database applications are described herein. Changes made to a first entry in a database table are recorded and an aggregate change is determined based at least in part on a data type of a second entry associated with the first entry. The second entry is then updated based at least in part on the aggregate change.

Abstract: Systems and methods are provided to implement an auto-partitioning second index (APSI) for a table in a storage service. In response to a creation request for the APSI that specifies a sort key but not a partition key for the APSI, the storage service automatically generates a partition key for the APSI, and allocates partitions for the APSI. When an item is stored in the APSI, a partition key value is auto-generated for the item, and the item is stored in a partition corresponding to its partition key value, and in a location within the partition according to its sort key value. When a query is received for the APSI, subqueries are sent to each partition, and the results from the subqueries are aggregated. When a repartition condition is detected, the APSI is repartitioned, reassigning the items in the APSI to a new set of partitions.

Abstract: Techniques described herein are directed to improved artificial neural network machine learning techniques that may be employed with a recommendation system to provide predictions with improved accuracy. In some embodiments, item consumption events may be identified for a plurality of users. From these item consumption events, a set of inputs and a set of outputs may be generated according to a data split. In some embodiments, the set of outputs (and potentially the set of inputs) may include item consumption events that are weighted according to a time-decay function. Once a set of inputs and a set of outputs are identified, they may be used to train a prediction model using an artificial neural network. The prediction model may then be used to identify predictions for a specific user based on user-specific item consumption event data.

Abstract: A method for writing data, the method may include: receiving or generating, by an interfacing module, a data unit coherent write request for performing a coherent write operation of a data unit to a first address; receiving, by the interfacing module and from a circuit that comprises a cache and a cache controller, a cache coherency indicator that indicates that a most updated version of the content stored at the first address is stored in the cache; and instructing, by the interfacing module, the cache controller to invalidate a cache line of the cache that stored the most updated version of the first address without sending the most updated version of the content stored at the first address from the cache to a memory module that differs from the cache if a length of the data unit equals a length of the cache line.

Abstract: The subject technology provides for receiving, via an application programming interface (API), a request from a client, the request including software configuration data for configuring data storage for supporting a workload of an application. A set of hardware performance characteristics for supporting the workload of the application is determined based at least in part on the software configuration data that includes a name of a respective software application and/or version of the software application. The subject technology determines a current set of available hardware resources that match the set of hardware performance characteristics. Using the current set of available hardware resources, a set of data stores are provisioned. The subject technology then provides, to the client, information related to a virtual mount point for provisioning the set of data stores to the application.

Abstract: A network visualization service may auto-generate graphical, dynamic, and interactive network diagrams of the infrastructure (resource instances, connections, etc.) of clients' virtual private networks as implemented on a provider network. A network diagram may include representations of various virtualized components of a client's virtual private network, as well as relationships among and connections between and among the various components. The diagram may also display logical and/or geographical groupings of the virtual resources in the client's virtual private network. The service may track changes to the virtual private network and update the diagram accordingly. The diagram may provide a user interface via which the client may select particular graphical objects on diagram to display additional information about a respective resource instance or connection and/or to change the client's virtual private network configuration via the network diagram.