Abstract: Systems, methods and computer-readable media are described for pre-warming compute instances in datacenters. A service provider associated with the datacenters may expect a demand for the compute instances and pre-configure computing resources within the datacenters to pre-launch the compute instances. As such, when a user requests a compute instance, the service provider may satisfy the request by allocating a pre-warmed compute instance to the user.

Abstract: An indirection mapping data structure can maintain a mapping between logical block addresses used by a host computer and physical data storage locations on an SSD. The indirection mapping data structure can include a existence map as well as a user data map. Trim data may be received from a host computer or other entity and may be used to mark particular logical addresses that are not in use as trimmed. To ensure trim data persists even if the indirection mapping data structure is lost, the existence map may be regularly stored to disk. The existence map can be journaled in the same way as user data stored to disk. When the indirection mapping data structure is rebuilt, the spatially coupled journals can be retrieved from the predefined scattered locations and used to identify the stored trim data which can be used to rebuild the existence map.

Abstract: Techniques for automated speech recognition (ASR) are described. A user can upload an audio file to a storage location. The user then provides the ASR service with a reference to the audio file. An ASR engine analyzes the audio file, using an acoustic model to divide the audio data into words, and a language model to identify the words spoken in the audio file. The acoustic model can be trained using audio sentence data, enabling the transcription service to accurately transcribe lengthy audio data. The results are punctuated and normalized, and the resulting transcript is returned to the user.

Abstract: Methods and systems for forecasting demand are described. A method may include determining a demand pattern for each respective item of at least some items of a plurality of items. The method may also include clustering the plurality of items into a plurality of clusters based on the determined demand patterns. The method may further include determining a composite demand pattern for each cluster. The method may additionally include forecasting a demand for an item of the plurality of items based on the composite demand pattern of the cluster to which the item belongs.

Abstract: Techniques for managing and discovering data using visual or audible representations may be provided. The data, in some examples, may correspond with individual items, and the system may correlate each item with a representation of the item (e.g., for navigation, for accessibility, etc.). The representations may be provided to allow discovery of the items in response to adjusting the resolution, audibly requesting information, or panning across a display to discover items that are displayed as these representations outside of the original presentation of representations. In some examples, the representation of the item may not be unique to the item or may be represented as a placeholder image or description. The higher the resolution, the more detail about the item may be provided. When an item is selected, item data may be provided that corresponds with the item.

Abstract: An electrowetting display device may comprise pixel regions arranged on a support plate, wherein each of the pixel regions includes a thin film transistor (TFT); a sensor disposed on the support plate, wherein the sensor is configured to generate a voltage based, at least in part, on light impinging on at least a portion of the pixel regions; and a reset control circuit configured to: based, at least in part, on the voltage generated by the sensor, determine an amount of modulation of the light; generate a reset pulse based, at least in part, on the determined amount of modulation of the light; and apply the reset pulse to each of the TFTs of at least some of the pixel regions.

Abstract: Audio/video (A/V) recording and communication devices in network communication with additional cameras in accordance with various embodiments of the present disclosure are provided. In one embodiment, an audio/video (A/V) recording and communication device is provided comprising: a first camera configured to capture image data at a first resolution; a communication module; and a processing module operatively connected to the first camera and the communication module, wherein the processing module is in network communication with a backend server, the processing module comprising: a processor; and a camera application that configures the processor to: maintain the first camera in a low-power state; receive a power-up command signal from the backend server based on an output signal from a second camera; power up the first camera in response to the power-up command signal; and capture image data using the first camera in response to the power-up command signal.

Abstract: Methods and apparatus for reconfiguring hosts in provider network environments in which hosts are evaluated to determine if steps of a full rebuild can be skipped. The hosts may implement slots of different types for virtual machines (VMs). Upon detecting that slots of a particular type are needed, a host that implements slots of another type may be selected for reconfiguration. The host may be evaluated to determine if one or more steps of a full rebuild can be skipped. The host may then be reconfigured to implement slots of the target type according to results of the evaluation. In at least some reconfigurations, at least one step of a full rebuild procedure is not performed for the respective host. Results of previous reconfigurations may be fed back into the evaluation process and used as one of the criteria for determining if steps can be skipped.

Abstract: A distributed data store may implement passive distribution encryption keys to enable access to encrypted data stored in the distributed data store. Keys to encrypt a data volume stored in the distributed data store may be encrypted according to a distribution key and provided to a client of the distributed data store. Storage nodes that maintain portions of the data volume may receive the encrypted key from a client to enable access to the data volume. The storage nodes may decrypt the key according to the distribution key and enable access to the data volume at the storage nodes. In some embodiments, a key hierarchy may be implemented to encrypt the keys that provide access to the encrypted data. The key hierarchy may include a user key.

Abstract: Methods, systems, and computer-readable media for capacity reservation for virtualized graphics processing are disclosed. A request is received to attach a virtual GPU to a virtual compute instance. The request comprises one or more constraints. Availability information is retrieved from a data store that indicates virtual GPUs available in a provider network and matching the one or more constraints. A virtual GPU is selected from among the available virtual GPUs in the availability information. The selected virtual GPU is reserved for attachment to the virtual compute instance. The virtual compute instance is implemented using CPU resources and memory resources of a physical compute instance, the virtual GPU is implemented using a physical GPU in the provider network, and the physical GPU is accessible to the physical compute instance over a network.

Abstract: When a client requests a data import job, a remote storage service provider provisions a shippable storage device that will be used to transfer client data from the client to the service provider for import. The service provider generates security information for the data import job, provisions the shippable storage device with the security information, and sends the shippable storage device to the client. The service provider also sends client-keys to the client, separate from the shippable storage device (e.g., via a network). The client receives the device, encrypts the client data and keys, transfers the encrypted data and keys onto the device, and ships it back to the service provider. The remote storage service provider authenticates the storage device, decrypts client-generated keys using the client-keys stored at the storage service provider, decrypts the data using the decrypted client-side generated keys, and imports the decrypted data.

Abstract: Methods and apparatus that allow clients to connect resource instances to virtual networks in provider network environments via private IP. Via private IP linking methods and apparatus, a client of a provider network can establish private IP communications between the client's resource instances on the provider network and the client's resource instances provisioned in the client's virtual network via links from the private IP address space of the virtual network to the private IP address space of the provider network. The provider network client resource instances remain part of the client's provider network implementation and may thus also communicate with other resource instances on the provider network and/or with entities on external networks via public IP while communicating with the virtual network resource instances via private IP.

Abstract: A speech recognition platform configured to receive an audio signal that includes speech from a user and perform automatic speech recognition (ASR) on the audio signal to identify ASR results. The platform may identify: (i) a domain of a voice command within the speech based on the ASR results and based on context information associated with the speech or the user, and (ii) an intent of the voice command. In response to identifying the intent, the platform may perform multiple actions corresponding to this intent. The platform may select a target action to perform, and may engage in a back-and-forth dialog to obtain information for completing the target action. The action may include streaming audio to the device, setting a reminder for the user, purchasing an item on behalf of the user, making a reservation for the user or launching an application for the user.

Abstract: A distributed execution environment can provide access to field-programmable device resources. The field-programmable device resources can be provided in association with one or more instances that are instantiated within the distributed execution environment upon request from a computing system. The computing system can be associated with a customer of the distributed execution environment. The customer can program the field-programmable device resources using designs created by or for the customer.

Abstract: A data storage system includes multiple head nodes and data storage sleds. The data storage sleds include multiple mass storage devices and a sled controller. Respective ones of the head nodes are configured to obtain credentials for accessing particular portions of the mass storage devices of the data storage sleds. A sled controller of a data storage sled determines whether a head node attempting to perform a write on a mass storage device of a data storage sled that includes the sled controller is presenting with the write request a valid credential for accessing the mass storage devices of the data storage sled. If the credentials are valid, the sled controller causes the write to be performed and if the credentials are invalid, the sled controller returns a message to the head node indicating that it has been fenced off from the mass storage device.

Abstract: A method of route discovery in wireless mesh networks comprises: receiving, using a first radio channel, by a first mesh network device of a wireless mesh network, a first route request identifying a destination mesh network device of the wireless mesh network, a first count of interfering communication links on a network path traversed by the first route request, and a second radio channel utilized by a second mesh network device along the network path; determining that the first radio channel matches the second radio channel; incrementing the first count of interfering communication links to produce a second count of interfering communication links; and broadcasting a second route request including an identifier of the destination mesh network device, a value reflecting the second count of interfering communication links, and an identifier of the first radio channel.

Abstract: A system includes access controls that allow or deny a request based at least in part on a context associated with the request. A service receives a request from a client, and forwards information that describes the request and the context of the request to a context management service. The context management service determines whether the request should be allowed by querying a set of context validation plugins. Each context validation plugin analyzes particular characteristics of a request's context, and based at least in part on the request's context, indicates to the context management service that the request should be allowed or denied. If a quorum of the context validation plugins indicate to the context management service that the request should be allowed, the context management service indicates to the service that the request is allowed within the provided context.

Abstract: A system and method includes a first device (e.g., a primary wireless earbud) connected to a second device (e.g., a secondary wireless earbud) using a first wireless connection. The first device is also connected to a third device (e.g., a smartphone or smart watch) using a second wireless connection. If the first device determines that a handoff-triggering event has occurred, the first device disconnects the second wireless connection and the second device and the third device establish a third wireless connection between them.

Abstract: The present disclosure is directed to parallelization of artificial neural network processing by conditionally synchronizing, among multiple computer processors, either the input or output of individual operations, and by conditionally using either rows or columns of certain matrices used in the operations. The conditional processing may depend upon the relative sizes of the input and output of the specific operations to be performed. For example, if a current layer matrix of values is larger than a next layer matrix of values to be computed, then rows of a weight matrix may be used by the computer processors to compute the next layer matrix. If the current layer matrix is smaller than the next layer matrix, then columns of the weight matrix may be used by the computer processors to compute the next layer matrix.