99results about How to "Speed up reading" patented technology

The invention discloses a magnetic disk cache system based on a solid-state disk and belongs to the technical field of cache systems in computer data storage systems. The magnetic disk cache system based on the solid-state disk comprises a data block metadata maintenance module, a solid-state disk data replacement module and a dirty data block writing-back disk module. According to the magnetic disk cache system based on the solid-state disk, the reading and writing performance of disk data in a large-scale storage environment is improved through the solid-state disk, and the data block reading and writing hit rate and the cache space utilization rate are high; meanwhile, it is guaranteed that data blocks cached under the condition that a computer is restarted due to faults can not be lost easily, the cold-to-hot convergence time of solid-state disk equipment is shortened, and the problems that according to a computer data storage system in the prior art, the cache space is limited, cached data can be lost easily, and the cache space can not be fully utilized can be solved.

The invention relates to a GPU (graphic processing unit) based parallel power flow calculation system and method for a large-scale power system. The system comprises a symbol Jacobian matrix forming and decomposing module, an initialization module, a power flow equation right-hand side calculation module, a jacobian matrix assignment module, an LU decomposing module and a forward and backward substitution module; the symbol Jacobian matrix forming and decomposing module is located on a host side, and the host side transmits calculating data to an equipment side; the power flow equation right-hand side calculation module, the jacobian matrix assignment module, the LU decomposing module and the forward and backward substitution module are sequentially connected on the equipment side. The method includes (1) transmitting data needed by calculation to the host side entirely; (2) generating a symbol Jacobian matrix and performing symbol composition on the symbol Jacobian matrix; (3) transmitting a decomposition result by the host side to the equipment side; (4) executing power flow equation right-hand side calculation; (5) executing Jacobian matrix assignment; (6) executing LU decomposition; (7) executing forward and backward substitution.

The invention discloses a data read-write method based on a mixed storage device. The mixed storage device comprises a small non-volatile particle addressing medium and a large non-volatile particle addressing medium. An address mapping table is stored in the small non-volatile particle addressing medium and comprises at least one address mapping record, and each address mapping record comprises a logic address, a physical address corresponding to the logic address and the number of times of write operations corresponding to the logic address. The method comprises the step of receiving an operation request, the step of analyzing the logic address of the operation request and determining the physical address of data corresponding to the operation request in the mixed storage device according to the address mapping table, the step of reading and writing the data at the corresponding position in the corresponding storage medium according to the physical address. The invention further provides a corresponding data read-write device. By means of the data read-write method and device, the extra write expenditure can be reduced, the read-write efficiency can be improved, and the service life of the storage device can be prolonged.

The present invention relates to a high-capacity FLASH solid-memory controller, and comprises a data cache unit, a checking code cache unit, a checking code generating and correcting unit, a bad block auto-processing unit and a dual cache control unit. The checking code generating and correcting unit is used for receiving the data transmitted from an external microcontroller through the microcontroller interface, generating the checking code according to the received data, and transmitting the data and the checking code to the dual cache control unit. The data transmitted from the dual cache control unit are corrected and are transmitted to the external microcontroller through the microcontroller interface. The bad block auto-processing unit is used for automatically identifying the bad memory block in an external FLASH memory, and distributing the replacement block, so as to guarantee the regular reading and writing of the external FLASH memory. The high-capacity FLASH solid-memory controller can be operated without the user. The dual cache control unit is communicated with the microcontroller interface, the FLASH interface or the bad block auto-processing unit, so as to realize the data writing or reading of the data cache unit and the checking code cache unit.

The invention discloses a full-memory operation system which supports container running. The full-memory operation system comprises a full-memory operation system body, and the full-memory operation system body comprises a generating unit and an operation system loading unit; the full-memory operation system generating unit is used for establishing a Linux customizing operation system, compiling tool package software and a container engine and establishing a Linux memory operation system mirror image; the operation system loading unit is used for achieving core module loading, storage driving loading, container mirror image configuration loading and container storage driving loading of the operation system. The invention provides a memory operation system and an achieving method thereof which support container application deployment, and are high in practicability and efficiency. According to the operating system, a state-free container can totally run in a memory, therefore the bottleneck of hardware reading and writing is avoided, reading of the container mirror image is accelerated, the situation that the container state is stored in persistent data is supported, the situation that a host machine or the network stores persistent data is further supported, and the data safety is guaranteed.

The invention provides an orderly table data management method based on a Hadoop distributed file system (HDFS) and a method and a system for reading table data. The orderly table data management method based on the HDFS comprises the following steps of: receiving the name of a to-be-operated table input by a user and bulk data comprising multiple operation records; sorting the received bulk data based on Hadoop, generating index data from the sorted bulk data, storing the sorted bulk data and the generated index data in a file form under a specified directory of the HDFS, and transmitting the name of the table, the file name of a file stored with the bulk data, the file name of a file stored with the index data and path data of the specified directory to a master server.

The invention provides a multi-application-system unified login method. The method comprises the following steps that a server establishes a user information view through Mysql; the server receives user information sent by a client, and generates a login key after verification is successful; the login key is carried in an application system access address and the application system access addressis sent; the client receives the application system access address and accesses an application system; the server determines whether an access source is safe and whether the login key is valid, and after the verification is successful, the application system establishes a login session credential based on the user information; and the client logs in the application system through the session credential. The invention also provides a multi-application-system unified login device. In the invention, through a same client, the different application systems can be logged in, and the different kindsof user identity verifications are integrated and unified into the same server so that a user does not need to remember the login addresses of the plurality of application systems and the safety of the login key is greatly increased.

The invention relates to a flexible log storage and acquisition method, which comprises a data writing process and a data reading process. The method is based on multi-file hash, and under the condition of single file reading-writing entry when a plurality of files are required to be frequently read and written, an index-based file storage and reading mode is provided, so that the files can be conveniently stored in a hash way, and required logs can be easily read.

The invention provides a three-dimensional memory and a three-dimensional memory manufacturing method. The three-dimensional memory comprises a substrate, a stack structure and an external structure,wherein the stack structure and the external structure are arranged on the substrate; the external structure is connected to the peripheral edge of the stack structure, and the external structure is provided with a conductive contact penetrating through the external structure; a channel structure penetrating through the stack structure is arranged in the stack structure; a conductor extending parallel to the substrate is arranged on one side, deviating from the stack structure, of the substrate, a first conductive part penetrating through the substrate is arranged at one end of the conductor,and the first conductive part is electrically connected with the conductive contact; and the other end of the conductor is provided with a second conductive part, and the second conductive part is embedded into the substrate corresponding to the stack structure and is electrically connected with the channel structure, so a common source contact in the stack structure in related technologies can becanceled, the problem of forming a coupling capacitor is avoided, and the reading and erasing rates of the three-dimensional memory are improved.

The invention discloses a storage system online data compression method, device and equipment, and the method comprises the steps: carrying out the data compression after receiving a compression instruction, and obtaining a compressed data block; Searching corresponding segment metadata according to the main metadata, determining the initial position of the current blank storage area according tothe segment metadata, and storing the compressed data block into a physical disk according to the initial position; After a reading instruction is received, searching corresponding segment metadata according to the main metadata and the reading position of the reading instruction; Determining the position of the data to be read in the physical disk according to the searched segment metadata, and then reading the data; Wherein the segment metadata comprises feature information of segments, and the segments are storage area segments obtained by dividing virtual storage addresses corresponding tothe physical disks according to a fixed length; Wherein the main metadata comprises feature information of the segment metadata. The metadata is segmented, each segment corresponds to the virtual storage address with the fixed length, reading and writing of the compressed data are facilitated, the compression performance is improved, and management of the compressed data is facilitated.

The invention provides a three-dimensional observation system surface element attribute multi-thread rapid display method, belonging to the seismic prospection and exploitation field. The method performs partitioning pretreatment on a three-dimensional observation system surface element attribute analysis result, and utilizes a multi-thread mode to increase covering frequency data reading and display drafting efficiency. The method comprises: (1) establishing a three-dimensional observation system covering frequency attribute matrix structure; (2) dividing the covering frequency attribute matrix structure into N sections in a surface element grid x-line direction; (3) obtaining the size of a current window drafting area, and creating N pieces of QPixmap pictures, wherein each picture corresponds to 1/N of the covering frequency attribute matrix structure; (4) creating N threads, respectively loading a covering frequency, and drafting the covering frequency into a corresponding QPixmap picture; and (5) drafting the N pieces of QPixmap pictures on corresponding positions of a screen. The method innovatively realizes three-dimensional observation system surface element attribute pretreatment.

The invention relates to a road traffic management system and a method. The road traffic management system comprises a second camera, a first camera, a converter, an identity reading apparatus, a radar speed measurement apparatus, a controller, a center server and a display. The radar speed measurement apparatus is used for vehicle speed measurement. After a vehicle is in an overspeed state, the controller makes the identity reading apparatus capture overspeed vehicle information and transmit to a memory in the center server. Simultaneously, after the vehicle is in the overspeed state, the controller outputs a overspeed prompt signal to the center server. The center server immediately intercepts a video signal of the first camera and stores in the memory of the center server, which is convenient for checking and peccancy processing.

The embodiment of the invention discloses a cache processing method and device. The method can comprise the following steps that: Cache receives a main memory address sent from a CPU (Central Processing Unit), wherein the main memory address comprises a Tag and an Index 1; and a Set1 corresponding to the Index1 is searched, and a Set2 corresponding to an Index2 is searched; and a matched Cache Line is searched in the Cache Line of the Set1 and the Set2. When the method is adopted, power consumption expenditures during Cache deficiency can be lowered, and Cache performance is improved.

The invention relates to the field of lithium ion battery prediction, and relates to estimation of the state of charge of a lithium ion battery based on an XGBoost model, wherein the estimation comprises the following steps: dividing discharge data of the lithium ion battery into a training data set and a test data set, and taking voltage, current and temperature in the data set of the training data set as features and inputting into the XGBoost model; setting parameters of the XGBoost model; training the training data set by using the XGBoost model; judging an error between a predicted SOC ofthe training data set and a real SOC of the training data set, and if the error is minimum, taking the set parameters of the XGBoost model as optimal parameters; taking the voltage, current and temperature in the test data set as features and inputting into the XGBoost model, and predicting the test data set by using the obtained optimal parameters of the XGBoost model to obtain a predicted SOC of the test data; and the estimation precision and robustness are improved.

The invention provides a configuration method, a receiving method, a terminal and network side equipment. The configuration method applied to the network side equipment comprises: sending first synchronous broadcast signal block measurement timing configuration SMTC configuration information, and the first SMTC configuration information being used by a terminal in an idle state or an inactive state. Thus, the terminal can read or measure the synchronous broadcast signal block of the first cell according to the received first SMTC configuration information of the first cell in an idle state oran inactive state, so that the speed of reading or measuring the synchronous broadcast signal block can be increased, the power consumption of the terminal can be saved, and the power consumption of the terminal can be reduced.

The invention discloses a data processing system in an environment with radio frequency identification tags densely arranged. The system is composed of an RFID reader, a rear-end data processing system, a reader device, an RFID tag module and a coupling element, wherein the a computer and a database are arranged in the rear-end data processing system, the reader device comprises hardware capable of sending energy of various light beams and sending and receiving information in a wireless mode, and a coil (the coil is an antenna when an RFID tag antenna is not integrated in the RFID tag module) is arranged in the coupling element. The data processing system has the advantages that an RFID reader holder can directly and fast read tag information from a large number of RFID tags through light starting, time is saved, and working efficiency is improved.

The invention relates to a three-dimensional image reconstruction method and system. The three-dimensional image reconstruction method comprises the following steps: pre-processing an original image to generate a first resolution two-dimensional image; re-sampling the first resolution two-dimensional image to generate a second resolution two-dimensional image of which the resolution is lower than that of the first resolution two-dimensional image; performing three-dimensional reconstruction on the sequences of the second resolution two-dimensional image to form a three-dimensional reconstruction image, and cutting an ROI (Region of Interest), in which a plurality of massive images constitute a three-dimensional reconstruction body, in the three-dimensional reconstruction image; calculating the coordinates, the width and the height of the ROI, and calculating the start frame and the terminate frame in the image sequences of the three-dimensional reconstruction body; processing the sequences of the first resolution two-dimensional image one by one according to the calculation results to obtain a plurality of massive image sequences; performing three-dimensional reconstruction on the massive image sequences to preview the ROI. The three-dimensional image reconstruction method and system can realize quick high-resolution previewing of images in the low-configuration environment.

The invention provides an automatic checking method for a transformer substation monitoring information acceptance card. A scanning thread reads the record of the alarm table in the database and records the last piece of information, when the next scanning thread runs for a period, the keyword of the last record is compared with the keyword of the last time, and when the keywords are different, the thread reads the information between the two records into the cache. newly-added records are read item by item, and whether the newly-added records are in the cached binary sorting tree or not is determined by calling a binary sorting tree searching method; And if the record exists in the binary sorting tree, the flag domain in the record value is set as 1, which indicates that the record is checked, thereby solving the problem that a large amount of time and manpower are consumed for the manual test of the master station and the station personnel in the current station-station joint debugging test work, and improving the working efficiency and the working accuracy. The function of automatically checking signals in an acceptance card of an existing power system is achieved, the operationindentifying level of the power system is improved, and the error occurrence probability is reduced.

The invention discloses a video-based fast template matching GPU implementation method which comprises the following steps: (1) copying image data to GPU equipment, wherein the operation of copying the images to the equipment and image template matching are executed at the same time as two independent streams, one of which executes a current-frame image template matching calculation process operation while the other of which executes a next-frame image copying operation; (2) storing the image template data in a shared memory, storing the image data in a texture memory, and randomly accessing by virtue of the texture memory for computation of match degree with the image template; (3) carrying out calculation at the same time by virtue of a multi-thread parallel calculation mode, wherein one thread calculates the template match degree quantity value of one position; and (4) determining the maximum value or the minimum value of the template match degree quantity value by virtue of a global atom and a shared atom so as to obtain a template matching result. According to the method disclosed by the invention, the template matching time can be obviously shortened and the practical value of a template matching algorithm can be improved.