Abstract: A parallel data processing method based on multiple graphic processing units (GPUs) is provided, including: creating, in a central processing unit (CPU), a plurality of worker threads for controlling a plurality of worker groups respectively, the worker groups including a plurality of GPUs; binding each worker thread to a corresponding GPU; loading one batch of training data from a nonvolatile memory to a GPU video memory corresponding to one worker group; transmitting, between a plurality of GPUs corresponding to one worker group, data required by data processing performed by the GPUs through peer to peer; and controlling the plurality of GPUs to perform data processing in parallel through the worker threads.

Abstract: Provided is a method for processing input data in a storage system and in communication with a repository. Views are generated that comprise a tree of nodes selected from a subset of nodes in a hierarchical representation of a schema. The views are saved to the repository. At least one of the views are used to create a job comprising a sequence of data transformation steps to transform the input data described by input schemas to the output data described by output schemas.

Abstract: A data deduplication method is executed by a controller for a solid state drive (SSD). The controller receives a signature for a block of data. The controller performs a comparison of the signature and information in a signature library and determines whether or not the signature matches the information. The controller sends a signal that indicates a result of the comparison. If the signature and the information match then the signal has a first value indicating that the block of data is already stored on the SSD; if the signature and the information do not match then the signal has a second value that is different than the first value.

Abstract: The present disclosure provides test structures, fabrication methods thereof and test methods thereof. An exemplary test structure includes a substrate having a to-be-tested region having at least one fin and a peripheral region having at least one fin surrounding the to-be-tested region; an insulation layer covering portions of side surfaces of the fins; at least one first gate structure covering side and top surfaces of the fin in the to-be-tested region; second gate structures covering side and top surfaces of the fins in the peripheral region; source/drain regions formed in portions of the fins between adjacent second gate structures and portions of the fins between the first gate structure and adjacent second gate structures; and a plurality of first conductive structures formed between adjacent second gate structures in the peripheral region. The plurality of first conductive structures cross over and are on source/drain regions of at least two fins.

Abstract: A content distribution system may include a headend server and media converters, the headend server being configured to distribute content items to user devices via the media converters and gateway devices. Each of the media converters may include a media converter cache and may be coupled to the headend server and a subset of the gateway devices, where each of the user devices is communicatively coupled to one of the gateway devices. The headend server may be further configured to determine one of the content items that is expected to be requested by a group of the user devices, determine one of the media converters that is coupled, via the subset of the gateway devices, to a largest number of user devices in the group of the user devices, and coordinate storing the one of the content items in the media converter cache of the one of the media converters.

Abstract: A system includes at least one processor and processes an ETL job. The system analyzes a specification of the ETL job including one or more functional expressions to load data from one or more source data stores, process the data in memory, and store the processed data to one or more target data stores. One or more data flows are produced from the specification based on the one or more functional expressions. The one or more data flows utilize in-memory distributed data sets generated to accommodate parallel processing for loading and processing the data. The one or more data flows are optimized to assign operations to be performed on the one or more source data stores. The optimized data flows are executed to load the data to the one or more target data stores in accordance with the specification. Present invention embodiments further include methods and computer program products.

Abstract: A single phase permanent magnet motor includes a stator and a rotor relative to the stator. The stator includes a stator core with a winding wound therearound. The stator core includes stacked first and second stator core laminations. The first stator core lamination includes a plurality of first pole shoes. The second stator core lamination includes a plurality of second pole shoes. The structures of the first pole shoes and the second pole shoes are different. The first pole shoes and the second pole shoes are stacked substantially along the axial direction of the motor. The present invention reduces vibration and noise caused by the use of a single type of core laminations in the prior art.

Abstract: A system includes at least one processor and processes an ETL job. The system analyzes a specification of the ETL job including one or more functional expressions to load data from one or more source data stores, process the data in memory, and store the processed data to one or more target data stores. One or more data flows are produced from the specification based on the one or more functional expressions. The one or more data flows utilize in-memory distributed data sets generated to accommodate parallel processing for loading and processing the data. The one or more data flows are optimized to assign operations to be performed on the one or more source data stores. The optimized data flows are executed to load the data to the one or more target data stores in accordance with the specification. Present invention embodiments further include methods and computer program products.

Abstract: The embodiments of the present disclosure disclose an ultrasound imaging method and system, the method may include transmitting a plurality of plane wave ultrasound beams to a scan target and acquiring corresponding plane wave echo signals; transmitting focused ultrasound beams to the scan target and acquiring corresponding focused beam echo signals; acquiring a plurality of velocity components of a target point in the scan target using the plane wave echo signals, and acquiring velocity vectors of the target point according to the plurality of velocity components; acquiring an ultrasound image of the scan target using the focused beam echo signals; and displaying the velocity vector and the ultrasound image.

Abstract: A single phase permanent magnet motor includes a stator and a rotor rotatable relative to the stator. The stator includes a stator core and a winding wound around the stator core. The rotor includes a rotor core and a plurality of permanent magnetic poles. The stator core includes at least one first stator core lamination and at least one second stator core lamination stacked with the first stator core lamination. Inner structures of the first stator core lamination and the second stator core lamination are different. This structure reduces motor's vibration and noise, avoids possible startup dead point and improves the motor startup reliability.

Abstract: System, method, and computer program product to process parallel computing tasks on a distributed computing system, by computing an execution plan for a parallel computing job to be executed on the distributed computing system, the distributed computing system comprising a plurality of compute nodes, generating, based on the execution plan, an ordered set of tasks, the ordered set of tasks comprising: (i) configuration tasks, and (ii) execution tasks for executing the parallel computing job on the distributed computing system, and launching a distributed computing application to assign the tasks of the ordered set of tasks to the plurality of compute nodes to execute the parallel computing job on the distributed computing system.

Abstract: Provided are a computer program product, system, and method for generating a view for a schema. A schema provides a definition of elements, wherein at least one of the elements comprises a recursive element of a recursive data type, wherein the recursive data type allows for instances of the recursive element to include instances of that same recursive data type at different levels in a hierarchy. Indication is received to transform the recursive types to non-recursive structures. A view for the schema is generated including information on the indication to transform the recursive types to non-recursive structure, wherein the processing of the data defined by the view causes a parser to generate a structured element for each instance of the recursive element in schema data being parsed, wherein the generated structured element includes a unique identifier of the instance of the recursive element.

Abstract: A motor assembly includes a single phase motor and a friction clutch. The single phase motor includes a stator and a rotor. The friction clutch includes a pressing member rotating with the rotor, and a connecting member to be connected to a load. wherein when the pressing member rotates, the pressing member generates an axial or radial movement to thereby exert a pressing force to the connecting member to cause the connecting member to rotate due to a frictional force generated between the pressing member and the connecting member. The present invention further provides a fluid driving device incorporating the motor assembly.

Abstract: A single phase permanent magnet motor includes a stator core, windings wound around the stator core, and a permanent magnet rotor. The stator core includes an end portion and two spaced arm portions. Each arm portion includes a connecting arm connected to the end portion and a pole claw formed at a distal end of the connecting arm. The two pole claws define a space for receiving the rotor. The pole claws surround the space and form an arc pole surface. The arc pole surface is recessed to form a startup groove located at a central axis of the pole claw. The two pole claws are spaced apart to form two slot openings there between. A line connecting the two slot openings is orthogonal to the central axis of the pole claw.

Abstract: A single phase permanent magnet motor includes a stator core and a permanent magnet rotor rotatably disposed in the stator core. The stator core includes an end portion and a pole arm extending from the end portion. The pole arm includes two connecting arms connected to the end portion and two pole claws formed at distal ends of the connecting arms. The permanent magnet rotor is disposed between the two pole claws. The rotor comprises a rotor core and a plurality of permanent magnets embedded in the rotor core, and the rotor core is made of a magnetic material.

Abstract: A single phase brushless motor and a power tool are provided. The single phase brushless motor includes a stator and a rotor. The stator includes a stator core and windings wound around the stator core. The stator core includes a yoke and at least two teeth. The tooth includes a tooth body and a tooth tip. The tooth tip includes first and second pole shoes. The two pole shoes of each tooth are symmetrical about a center line of the tooth body. Each tooth defines a positioning groove facing the rotor between the two pole shoes. Pole shoes of adjacent two of the at least two teeth are spaced apart by a slot opening. A width of the positioning groove is greater than a width of the slot opening. The peak value of the cogging torque of the motor is increased, and the motor has a large startup torque.

Abstract: Embodiments of the present invention disclose a resource allocation method, a transmission method, and apparatuses. The resource allocation method comprises: sending a scheduling instruction, wherein the scheduling instruction is used to instruct a user equipment to prepare for transmission; determining an available carrier in an unlicensed spectrum; and sending a resource allocation message in response to that an available carrier exists, wherein the resource allocation message is used to instruct to allocate the available carrier to the transmission of the user equipment. According to the method and apparatus of the embodiments of the present invention, after an available carrier is determined, a resource allocation message is sent again to explicitly inform a user equipment of the available transmission resource, thereby avoiding the problem of resource unavailability and/or resource waste caused by the unpredictability of whether an unlicensed spectrum is available.

Abstract: A single phase motor and an electrical device are provided. The single phase motor includes a stator and a rotor. The stator includes a stator core and a winding wound around the stator core. The stator core includes a yoke and two opposed pole portions. Each pole portion includes short and long pole shoes. The rotor is received in a space defined by the short and long pole shoes of the two pole portions. The short pole shoe of each pole portion and the long pole shoe of the opposite pole portion are located adjacent to each other and define a slot opening therebetween. Each pole portion includes the short pole shoe and long pole shoe, which provides the motor with different startup capabilities along the clockwise and counter-clockwise directions.

Abstract: The present disclosure provides FinFET structures and fabrication methods thereof. An exemplary fabrication method includes providing a substrate having a first region and a second region; forming a first well region in first region and a second well region in the second region; forming at least one first fin in the first region and at least one second fin in the second region; forming a first doped layer covering the first fin; forming a second doped layer covering the second fin; forming first doped sidewall spacers on side surfaces of the first fin and second doped sidewall spacers on side surfaces of the second fin by a mask-less etching process; and performing a thermal annealing process to the first doped sidewall spacers and the second doped sidewall spacers to form a third well region in the first fin and a fourth well region in the second fin, respectively.

Abstract: The present invention provides a single phase permanent magnet motor including a stator core and a permanent magnet rotor. The stator core includes an end portion and two arm portions extending from the end portion. Each arm portion includes a connecting arm connected to the end portion and a pole claw formed at a distal end of the connecting arm. The two pole claws defines a receiving space. The rotor is rotatably disposed in the receiving space of the stator core. The rotor includes a rotor core made of a magnetic material and at least one permanent magnet attached to the rotor core.