222results about How to "Reduce read" patented technology

A system and method of information retrieval and triage for information analysis provides an for interactive multi-dimensional and linked visual representation of information content and properties. A query interface plans and obtains result sets. A dimension interface specifies dimensions with which to categorize the result sets. Links among results of a result set or results of different sets are automatically generated for linked selection viewing. Entities may be extracted and viewed and entity relations determined to establish further links and dimensions. Properties encoded in representations (e.g. icons) of the results in the multidimensional views maximizes display density. Multiple queries may be performed and compared. An integrated browser component responsive to the links is provided for viewing documents. Documents and other information from the result set may be used in an analysis component providing a space for visual linking, to arrange the information in the space while maintaining links automatically.

A read element includes a magnetoresistive sensor having a side and an upper surface that defines an edge. An underlayer overlies the side of the magnetoresistive sensor, and a hard bias layer overlies at least part of the underlayer and defines a hard bias junction with the underlayer. A lead is formed atop the hard bias layer. The hard bias junction is recessed from the edge of the magnetoresistive sensor by a predetermined recess distance, to provide stability and sensitivity to the read element.

A non-volatile memory system, such as a flash EEPROM system, is disclosed to be divided into a plurality of blocks and each of the blocks into one or more pages, with sectors of data being stored therein that are of a different size than either the pages or blocks. One specific technique packs more sectors into a block than pages provided for that block. Error correction codes and other attribute data for a number of user data sectors are preferably stored together in different pages and blocks than the user data.

To provide a configuration including a fully differential amplifier in which decrease in a reading speed can be suppressed. A photoelectric conversion apparatus according to the present invention includes a pixel area where a plurality of pixels are arranged; an amplifier configured to amplify a signal from the pixel area; a plurality of signal paths for transmitting the signals from the pixel area to the amplifier. The amplifier is a fully differential amplifier which includes a plurality of input terminals including a first input terminal and a second input terminal to which the signals from the plurality of signal paths are supplied and a plurality of output terminals including a first output terminal and a second output terminal and the input terminals and the output terminals have no feedback path provided therebetween.

The invention discloses a convolutional neural network accelerator based on calculation optimization of an FPGA. The convolutional neural network accelerator comprises an AXI4 bus interface, a data cache region, a pre-fetched data region, a result cache region, a state controller and a PE array. The data cache region is used for caching feature map data, convolution kernel data and index values read from an external memory DDR through an AXI4 bus interface; The pre-fetched data area is used for pre-fetching feature map data needing to be input into the PE array in parallel from the feature mapsub-cache area; The result cache region is used for caching a calculation result of each row of PE; The state controller is used for controlling the working state of the accelerator to realize conversion between the working states; And the PE array is used for reading the data in the pre-fetched data area and the convolution kernel sub-cache area to carry out convolution operation. The accelerator utilizes the characteristics of parameter sparsity, repeated weight data and an activation function Relu to end redundant calculation in advance, so that the calculation amount is reduced, and the energy consumption is reduced by reducing the access memory frequency.

A nonvolatile semiconductor memory device is provided in such a manner that a semiconductor layer is formed over a substrate, a charge accumulating layer is formed over the semiconductor layer with a first insulating layer interposed therebetween, and a gate electrode is provided over the charge accumulating layer with a second insulating layer interposed therebetween. The semiconductor layer includes a channel formation region provided in a region overlapping with the gate electrode, a first impurity region for forming a source region or drain region, which is provided to be adjacent to the channel formation region, and a second impurity region provided to be adjacent to the channel formation region and the first impurity region. A conductivity type of the first impurity region is different from that of the second impurity region.

A document reading apparatus is disclosed. The document reading apparatus includes a scanner; an energy-saving power supply unit; an energy-saving control unit; a clock IC; an output compensation unit for updating digital conversion parameters including image signal amplification gain such that image data of a reference white board read by a CCD of the scanner are made into a proper value; and an output compensation controlling unit. The output compensation controlling unit reads time data when an operation mode is shifted from pause mode to waiting mode, stores the digital conversion parameters updated by the output compensation unit in a non-volatile memory, updates operation time with the present time if the elapsed time from a previous operation time stored in the non-volatile memory is equal to or greater than a setup value, and uses the digital conversion parameters stored in the non-volatile memory as they are, if the elapsed time from the previous operation time stored in the non-volatile memory is less than the setup value.

Disclosed is a data management apparatus comprising a message transfer/analysis unit that receives an object data access request from a client and returns a response; an object metadata storage unit that stores, as metadata, a set of ancillary information on data for an object and a data area corresponding to the ancillary information; an object metadata management unit that performs reading, updating, and registration processing for the metadata; a data conversion unit that sends the data access request to a data access unit, extracts object data; and a data access unit that sends a data access command, specified by the data conversion unit, to a file server or a database server and sends response data, received from the file server or the database server, to the data conversion unit.

A nonvolatile semiconductor memory device comprises a memory cell including a variable resistance element changing its electric resistance by voltage application and having current-voltage characteristics in which a positive bias current flowing when a positive voltage is applied from one electrode as a reference electrode to the other electrode through an incorporated rectifier junction is larger than a negative bias current, a memory cell selection circuit for selecting the memory cell from the memory cell array, a voltage supply circuit for supplying a voltage to the memory cell so that a predetermined positive voltage corresponding to the reading operation is applied to the other electrode of the variable resistance element, in the reading operation, and a readout circuit for detecting the amount of the positive bias current and reading the information stored in the selected memory cell, in order to suppress the reading disturbance of the memory cell.

An MPEG video decoder 1 decodes a MPEG video stream using a discrete cosine transform together with a motion compensated prediction performing backward prediction and forward prediction. A frame buffer 104a is provided with a storage area for forward-reference luminance data used for the backward prediction and a storage area for rearward-reference color-difference data used for the forward prediction. A frame buffer 104b is provided with a storage area for forward-reference color-difference data used for the backward prediction and a storage area for rearward-reference luminance data used for the forward prediction. Memory access for each of the frame buffers 104a, 104b with an input/output data bus width of 16 bit is performed in a parallel processing.

Patterned, longitudinally and transversely antiferromagnetically exchange biased GMR sensors are provided which have narrow effective trackwidths and reduced side reading. The exchange biasing significantly reduces signals produced by the portion of the ferromagnetic free layer that is underneath the conducting leads while still providing a strong pinning field to maintain sensor stability. In the case of the transversely biased sensor, the magnetization of the free and biasing layers in the same direction as the pinned layer simplifies the fabrication process and permits the formation of thinner leads by eliminating the necessity for current shunting.

The present invention relates to a method and a system for detecting presence in a predefined space. The system comprises cascade connected sensors (s₁, s₂, . . . , s_n), an output device (1), a control unit (2) and a processing unit (3). The method comprises the steps of activating a first sensor (s₁) in the cascade and waiting until the first sensor detects presence. When presence is detected, a successive sensor (s₂, . . . , s_n) in the cascade is activated, and when the successive sensor (s₂, . . . , s_n) also detects presence the step of activating the successive sensor is repeated until all sensors in the cascade have been activated. If the successive sensor (s₂, . . . , s_n) does not detect presence the method returns to the waiting step of the first sensor. When the ultimate sensor in the cascade (s_n) detects presence, a confirmation message is sent to the output device (1).

The invention discloses an implementation method for operator reuse in a parallel database, comprising the following steps of: step 1, generating a serial query plan for query through a normal query planning method, wherein the query plan is a binary tree structure; step 2, executing the query plane by scanning from top to bottom, searching materialized reusable operators, changing the query plane structure, and changing thread level materialized operators into global reusable materialized operators; step 3, parallelizing the query plan changed in the step 2, and generating a plan forest for parallel execution of a plurality of threads; step 4, executing global reusable operator combination on the plan forest generated in the step 3, and generating a directed graph plan for the materialized reusable operators capable of being executed by the plurality of threads in parallel; step 5, executing own plan part in the directed graph by each thread in parallel, wherein the thread which executes the global reusable operator firstly is called a main thread, the main thread locks the global reusable operator and truly executes the operator and the plan of the operator, and other threads wait; step 6, unlocking the global reusable operator by the main thread after execution, wherein other threads start to read data from the global reusable operator and continue to execute own plan tree;and step 7, releasing the materialized data of the operator by the main thread after all the plans read the data of the global reusable operator.

A digital copying device, which serves as an image reading apparatus for improving processing speed by adjusting the timing of various operations when reading is performed using a sheet-through system, causes a first traveling body to move beneath a white reference plate in order to read the white reference plate, and then performs an ADF shading operation to move the first traveling body beneath a sheet-through contact glass. The sheet-through contact glass is then subjected to a dirt detection operation by reading a dirt detection reference plate, whereupon an original is conveyed onto the sheet-through contact glass by an ADF to perform a sheet-through reading operation whereby an image of the original is read. When a sheet-through reading operation is to be performed, the sheet-through reading operation is started following the reception of an image reading request and after the elapse of a time (T=Ts+Tl−Td) obtained by subtracting a time Td, which extends from the beginning of the sheet-through reading operation to the point at which the tip of the original reaches a sheet-through reading position, from the sum of a time Ts required for the ADF shading operation and a time Tl required for the dirt detection operation.

An assay apparatus having a sample vessel within which an assay may be performed. The apparatus further includes a holder having a receptacle, socket or other device configured to operatively receive the sample vessel in a precise and easily repeated location with respect to the holder. A magnet may be operatively associated with the holder such that a magnetic field generated by the magnet intersects a portion of the sample vessel defining a magnetic concentration region within the sample vessel. A separate or integrated detection or interrogation instrument, typically a spectrometer, may be provided.

A An information processing apparatus for processing data using a main memory device and a nonvolatile secondary storage device includes a nonvolatile main memory unit, a volatile main memory unit, a determination unit that determines whether the data is designated as confidential data, and a control unit that stores the data in the volatile main memory unit if the determination unit determines that the data is designated as confidential data and stores the data in the nonvolatile main memory unit if the determination unit determines that the data is not designated as confidential data.

A method is provided for forming a CPP MTJ or GMR read sensor in which the free layer is self-stabilized by a magnetization in a circumferential vortex configuration. This magnetization permits the pinned layer to be magnetized in a direction parallel to the ABS plane, which thereby makes the pinned layer directionally more stable as well. The lack of lateral horizontal bias layers or in-stack biasing allows the formation of closely configured shields, thereby providing protection against side-reading. The vortex magnetization is accomplished by first magnetizing the free layer in a uniform vertical field, then applying a vertical current while the field is still present.

The invention provides a data storage method and device, and a storage medium. The method comprises the steps of receiving a data storage request sent by a client, wherein the data storage request comprises to-be-stored data; determining a state of the to-be-stored data according to LFU and LRU of the to-be-stored data, wherein the state is a cold data state or a hot data state; and according to the state of the to-be-stored data, determining the storage medium corresponding to the to-be-stored data, and storing the to-be-stored data in the storage medium. According to the data storage methodand device, and the storage medium provided by the invention, the probability of reading a magnetic disk can be reduced, and the stability of a system is improved.