508results about How to "Improve reading and writing efficiency" patented technology

The invention relates to the technical field of network communications, in particular to a cluster communication system and method for improving message transfer performances. The system comprises a message release end, a message server cluster containing a plurality of message servers, a message subscribing end and a distributed coordination service cluster. The cluster communication system and method for improving message transfer performances have a global single point-free clustering expandable distributed design; if one or more nodes of the message release end or subscribing end have a fault, other nodes in the same group can continue to send or receive messages, and a processing flow of the messages is not interrupted; by adopting an index-based distributed file storage solution, the defects of the existing DB and file storage are effectively avoided, so that the read and write efficiency of the messages is higher; and by using a long poll PULL message transfer manner, the throughput is ensured while the message instantaneity is ensured.

A spin-transfer MRAM is described that has two sub-cells each having a conductive spacer between an upper CPP cell and a lower MTJ cell. The two conductive spacers in each bit cell are linked by a transistor which is controlled by a write word line. The two CPP cells in each bit cell have different resistance states and the MTJ cell and CPP cell in each sub-cell have different resistance states. The MTJ free layer rotates in response to switching in the CPP free layer because of a large demagnetization field exerted by the CPP free layer. An improved circuit design is disclosed that enables a faster and more reliable read process since the reference is a second MTJ within the same bit cell. When RMTJ1>RMTJ2, the bit cell has a “0” state, and when RMTJ1<RMTJ2, the bit cell has a “1” state.

A memory includes a nonvolatile memory cell array, and a memory control circuit which carries out writing of data to and reading of data from the memory cell array in access units of N bits where N is an integer equal to 2 or greater. The memory cell array includes a rewritable area in which both writing of data and reading of data are permissible, and a read-only area in which writing of data is prohibited and reading of data is permissible. The rewritable area is configured so that the N bits constituting one access unit contain both actual data and an error detection code. The read-only area is divided between an actual data area in which the N bits constituting one access unit contain actual data, and an error detection code area in which the N bits constituting one access unit contain error detection codes.

A spin-transfer MRAM is described that has two sub-cells each having a conductive spacer between an upper CPP cell and a lower MTJ cell. The two conductive spacers in each bit cell are linked by a transistor which is controlled by a write word line. The two CPP cells in each bit cell have different resistance states and the MTJ cell and CPP cell in each sub-cell have different resistance states. The MTJ free layer rotates in response to switching in the CPP free layer because of a large demagnetization field exerted by the CPP free layer. An improved circuit design is disclosed that enables a faster and more reliable read process since the reference is a second MTJ within the same bit cell. When RMTJ1>RMTJ2, the bit cell has a "0" state, and when RMTJ1<RMTJ2, the bit cell has a "1" state.

The invention discloses data writing and reading method and device in a distributed caching system. According to the distributed caching system, the indexing information of the data is stored in an internal memory, and the data information is stored in a hard disk; and when the writing and reading of the data are carried out, corresponding operation be performed on the hard disk through the indexing information stored in the internal memory. According to the method and device provided by the embodiment of the invention, the data is written and read in combination of the internal memory and the hard disk, and the internal memory is only used for storing indexing information; the reading and writing of the data are carried out in combination of the internal memory and the hard disk, so that the problem that all data is stored into the internal memory can be avoided, and the requirements on caching of a large number of data can be met; and furthermore, the data read-write efficiency can be improved by reading and writing the data in the combination of the internal memory and the hard disk, so the reliability of the system can be improved.

The invention relates to a multi-core processor and a multi-core processor set. The multi-core processor comprises at least one master core, at least one slave array, a first interconnecting structure and a slave core interconnecting structure, wherein the slave core array comprises a plurality of slave cores which are heterogeneous with the master core; the first interconnecting structure and the slave core interconnecting structure are used for communication between the master core and the slave core array; the slave core interconnecting structure is used for communication between any two slave cores in the slave core array; and as an optimized technical scheme, the master core is a universal processor core, the slave cores are processor cores microstructures and instruction sets of which are simplified and optimized, and the multi-core processor is integrated on the same chip. According to the multi-core processor and the multi-core processor set disclosed by the invention, communication efficiency among the processor cores is improved, calculation density of the whole multi-core processor is increased and balance between universal control function and high calculation density is realized.

The invention discloses a blockchain-based medical electronic medical record distributed management system and a construction method thereof. According to the blockchain-based medical electronic medical record distributed management system and the construction method thereof, the diagnosis processes of patients, such as medical records and prescriptions, are written into a blockchain in the form of smart contracts, so that a relevant hospital does not need to rely on a centralized server to manage the medical record information of the patients, and single-point failures common to the centralized server can be avoided; the data stored in the blockchain cannot be tempered and are traceable, and therefore, illegal tampering can be prevented; and a series of information such as the physical health, medical histories, and personal data of the patients, which concern personal privacy, can be effectively protected through an ABE encryption algorithm. With the system and method of the invention adopted, data loss caused by the single-point failures can be avoided, system maintenance cost can be decreased, and a condition that the diagnosis records of the patients are falsified during medical disputes can be avoided.

The invention is suitable for the technical field of data storage, and provides a data reading and writing method and device for flash memory equipment and a host system. The method comprises the following steps: applying for a plurality of reading and writing caches of first fixed values in the cache of the host system, wherein the first fixed values are integer times of flash memory pages in the flash memory equipment; during data writing, writing data of which the writing address are unaligned with the reading and writing caches and / or small file data and / or data of which the writing addresses are writing hot addresses in data to be written into the reading and writing caches; and when the number of empty reading and writing caches is smaller than or equal to a preset fourth fixed value and / or the host system is in an idle state and / or satisfies a preset condition, writing data in the reading and writing caches into the flash memory equipment. according to the method, the device and the system, the problems of data copy and migration performed continuously in a writing process due to misalignment of addresses or writing of the small file data can be reduced greatly, and the data reading and writing efficiency of the flash memory equipment is increased.

The invention discloses a flash memory storage system and a reading, writing and deleting method thereof. The flash memory storage system comprises a cache, a main control module, a cache metadata record sheet, a read mapping table and a write mapping table. The write mapping table is used for being stored in the cache and writing in the correspondence of a logical storage block and a physical storage block. The read mapping table is used for being stored in the cache and reading out the correspondence of the logical storage block and the physical storage block. The cache metadata record sheet is used for storing correspondences of metadata sheet addresses, the physical storage blocks and rear end flash memory addresses. By means of the flash memory storage system, the unnecessary write-in or read-out of a rear end flash memory can be reduced, the zero copy on a read-write data channel is achieved, the unnecessary intermediate copy process is omitted, and therefore the read-write efficiency is improved; the size of the read-write access of a front end application can be matched with the size of the rear end flash memory.

The invention discloses a method and device for data reading and writing on the basis of key-value buffer. The method includes: setting a primary buffer and secondary buffers buffering data in a key-value mode, setting that secondary buffers obtain synchronous updating data from the primary buffer according to a set synchronous cycle; receiving data requests sent by clients, outputting the data requests to the selected secondary buffer if the data requests are data reading requests, otherwise outputting the data requests to the primary buffer; receiving data request responses retuned from the primary buffer or data reading request responses returned from the primary buffers, and forwarding the data request responses and the data reading request responses to the corresponding clients; and monitoring state of the primary buffer and the secondary buffers, selecting the secondary buffers to send property change instructions after faults of the primary buffer are monitored, and switching the secondary buffers to the primary buffer according to the property change instructions. According to the method and device for the data reading and writing on the basis of the key-value buffer, the clients can use the buffers transparently, do not need to care about whether a single buffer breaks down, and can improve data reading and writing efficiency of the buffers.

The invention discloses a method of implementing configuration and management of data of a home gateway based on an XML (Extensible Markup Language). The method comprises the following steps of: respectively defining two data structures of TR-object and TR-param in an internal storage; loading and traversing an XML configuration file of the home gateway, configuring objects and parameters in the XML configuration file and the respective attribute information into the data structures of TR-object and TR-param in the internal storage, and constructing an internal storage node parameter tree; writing the configuration data in the internal storage node parameter tree into an FLASH of the home gateway and storing to form a new XML configuration file; and after restarting the home gateway, rereading the new XML configuration file in the FLASH and carrying out configuration and management. According to the method, an XML file used for unified storage of the data of the home gateway is designed by utilizing an XML rule, and when the parameter information changes, the configuration information is stored in the FLASH immediately, so that a configuration data loading process is simplified, the configuration data reading and writing efficiency is improved and the occupancy of the internal storage of a program is reduced.

The invention relates to the technical field of disk caching, in particularly to an object caching method based on a disk. The object caching method includes the steps that file storage space is divided to construct a file storage structure of secondary indexes, and the processes of adding objects to a disk cache and acquiring and deleting cached objects from the disk cache, a cache failure mechanism and the cache optimization arrangement process are achieved according to the file storage structure of the secondary indexes. Data of the objects are stored through chunks with the fixed sizes, the multiple small objects are merged and stored in one chunk, large objects are divided into a plurality of object blocks, the object blocks are stored in the multiple chunks respectively, the objects of any size and any type can be cached, and caching efficiency is higher.

The invention discloses an intelligent optical fiber wiring device and method and a management system. A machine frame electronic label is arranged on optical fiber wiring device equipment, a machine disc electronic label is arranged on each optical fiber wiring disc, and a connector electronic label is connected on each optical fiber movable connector. Meanwhile, an optical fiber wiring electronic label card reader is adopted, the optical fiber wiring electronic label card reader is provided with a plurality of electronic label card reading terminals which correspond to optical fiber wiring ports on the optical fiber wiring discs or optical fiber wiring ports on optical fiber wiring port sets one to one, the connector electronic labels on the current optical fiber wiring disc or the current port set and the electronic labels on the machine discs or the optical fiber wiring port sets can be read at a time. The scheme has high management efficiency and less labor misreading management loopholes, the wiring maintenance construction can be indicated and guided through the state of the optical fiber wiring ports, no electronic label read-write and management circuit needs to the added to the device, and a new intelligent ODN convenient to use can be constructed and intelligentized reconstruction of stock routine ODN equipment can be achieved.

The invention brings forward a data query method and system based on Key-Value data blocks. The method comprises following steps: selecting dimensions of data aggregation based on attributes of perception data, clustering perception data of the same attribute dimension in common, saving perception data to a same data block and utilizing selected data dimensions as attribute information corresponding to data blocks; adopting Key-Value data format to save and manage data in the data blocks; and saving and managing storage positions and corresponding attribute information by adopting a relational database. The data query method and system based on Key-Value data blocks has following beneficial effects: a deluge of perception data on the Internet of Things can be read and written in an efficient manner without using a great number of computers; and the data block is easy to expand according to business scale.

Disclosed is a spin hall effect-based magnetic random access memory. The magnetic random access memory is characterized in that a non-magnetic heavy metal layer, an MTJ unit and a side electrode are all cuboid columnar and are perpendicular to a circular crystal substrate surface; an upper electrode, a bottom electrode and an insulation layer are parallel to the circular crystal substrate surface;meanwhile, the free layer of the MTJ unit is in direct contact with the non-magnetic heavy metal layer; the MTJ unit and the side electrode are isolated from the upper electrode and the bottom electrode through the insulation layer; and the non-magnetic heavy metal layer is in direct contact with the upper electrode and the bottom electrode. A spin hall effect-based SOT-MARM storage unit is characterized in that when a current flows through the non-magnetic heavy metal layer, a spin current is generated based on the spin hall effect; the spin current flows through the free layer to be acted on a free layer magnetic torque in a torque form to be subjected to turnover so as to complete information writing in; and in addition, when the current flows through the MTJ unit and flows to the sideelectrode through the bottom electrode, information reading in the storage unit is completed.

The system is composed of dynamic executive control module, case scheduler module, module of syntax analysis, file interface, interface for testing executive software, and interface for testing management tool. The system converts relevant testing case to script of commercial testing software. Basic operations are encapsulated as a process so as to form structured method for calling processes from case to be tested. Using dynamic implementation of testing case and technique of delayed loading, the invention broken up the procedure for testing case into to two steps: 'initialization step of case test' and 'executive step of case test'. Thus, the invention possesses advantages: independence for editing case, reuse of cases and possesses, lowered occupancy rate of system memory and raised efficiency for executing cases.

The invention relates to a method and a device for utilizing an external memory to realize convolutional interleaving and deinterleaving. The method of the invention comprises the following steps that: interleaving and deinterleaving data is cached in an on chip buffer; when the on chip buffer is written or read, the on chip buffer is communicated with a storage controller; the interleaving / deinterleaving data is written in an external memory through the storage controller or is read out from the external memory. The method and the device utilize the on chip buffer to cache the interleaving / deinterleaving data, thereby the read-write efficiency on the external memory is greatly improved and the demand on read-write clock frequency of the external memory is reduced; the external memory is operated by a special external interface through the storage controller, thereby complex read-write time sequence can be screened and multiplexing of the external memory with other functional modules is also convenient.

The invention relates to a distributed file system which comprises a big file storage server, a big file metadata management server and a cache server, wherein the big file storage server is used for storing split big file data blocks, and big files are files of which the size is larger than the preset size; the big file metadata management server is used for storing metadata of the big files, storing mapping information of the big file data blocks on the big file storage server, managing the namespace of the big files and processing the requesting information of a user; the cache server is used for storing small files, metadata of the small files and caching part of frequently visited big files, and the small files are the files of which the size is smaller than or equal to the preset size. According to the distributed file system, the big files and the small files are stored separately, the big files are stored on the big file storage server in a block mode, the small files are stored on the cache server and the reading and writing efficiency of the big files and the small files is improved effectively.

The invention discloses a method, device and system for carrying out data reading and writing on the basis of a DDR SDRAN (Double Data Rate Synchronous Dynamic Random Access Memory), which are used for solving the problems that the efficiency is low and the cache utilization rate is poorly increased during reading-writing based on the DDR. The method comprises the steps of: according to the size of a chain table node, dividing a received message into a plurality of data blocks with corresponding lengths; aiming at each data block, dividing the data block into a plurality of data sub units, when the quantity of bytes included by the last data sub unit in the data block is not more than the length of a byte processed by one DDR every time; and splicing data of corresponding bytes in the last data sub unit and the first data sub unit in the next data block, and writing into the group of DDR SDRAM. Because when reading-writing operations are carried out through each group of DDR in the embodiment of the invention, in order to avoid the waste of DDR resources, the data is spliced, therefore, the efficiency of DDR reading-writing is increased, and the utilization rate of cache efficiency is increased.

A slave and a master of serial peripheral interface, system thereof and method thereof are provided. The invention saves the read address during the process of data-read in the serial peripheral interface (SPI) and adds a new data-read command so that when the former data address is followed by the next one, the transmission of the read address is ignored by using new read command in order to increase the efficiency of data read in the serial peripheral interface.

The invention relates to a data access and integration method and system, which can be applied in a data memory. The method comprises the following steps: receiving M data access commands with different types transmitted by a bus in order; when the M data access commands comprise N data access commands with the same command types and a plurality of access addresses which conform to continuation address relations, adjusting the orders of data access commands based on the orders of the access addresses and storing first data corresponding to N data access commands in the data memory. The invention can upgrade the read-write efficiency of the memory obviously.

The embodiment of the invention provides a method and system for reading and writing user data under virtual environment and a physical machine. The method includes the steps that whether every virtual machine has a user data reading and writing request or not is detected; the user data reading and writing requests of all the virtual machines are collected; the reading requests are added into a reading request queue, and the writing requests are added into a writing request queue; when the correspondence degree between first request addresses in the reading request queue and / or the writing request queue and second request addresses in the reading request queue and / or the writing request queue meets a preset condition, the request queue corresponding to the first request address is adjusted according to a request queue adjustment method corresponding to the preset condition; a request in the front of the request queue with the high priority is called and sent to a server where the user data are stored. By means of the method, reading and writing pressure of communication interfaces of the physical machine and the reading and writing pressure of the server where the user data are stored are substantially reduced, and user data reading and writing efficiency is improved.

The invention provides a method and a device for increasing the read-write rate of a double data rate synchronous dynamic random access memory (DDRSDRAM). The method comprises the following steps of: receiving a read address corresponding to a read request of a packet; caching the read address in a corresponding first first-in first-out queue (FIFO) according to a bank number; reading the read address out from a first FIFO in the first FIFO for the first time; if the next read address in the first FIFO and the read address read out for the first time belong to same-line addresses, continuously reading the first FIFO; and if the next read address in the first FIFO and the read address read out for the first time belong to different-line addresses, reading the next first FIFO in a polling mode. By the method and the device for increasing the read-write rate of the DDRSDRAM, realization difficulties are reduced, the realizability and the adjustment force of a field programmable gate array (FPGA) project are improved, and the read-write efficiency of the DDRSDRAM is greatly improved by limited resources.

The invention provides a data storage and cutting method and device and a storage medium. The method comprises the steps that father node information of all nodes except a root node of a first Merkeltree is stored in a local database; When the first leaf node is inserted into the first Merkel tree, storing a first index node corresponding to the first leaf node in a local database; Wherein the second key value of the first index node comprises the first key value of the first leaf node, the first block height and the first hash value; Wherein the second key value is used for searching a plurality of index nodes and corresponding leaf nodes which are the same as the first key value, and the father node information is used for searching redundant father nodes. According to the method, the index nodes corresponding to the leaf nodes are stored, redundant state data are deleted according to the index nodes, the data synchronization rate is greatly increased on the premise that the normalaccount state is not affected, and the read-write efficiency is improved.

The invention is a method for storing multiple virtual queues data based on FPGA. The invention is designed based on the technology realizing the network card data read-write in FPGA, in particular to a method of storing multiple virtual queues data, adopting a plurality of register blocks which contain a receiving engine module, a read-write module and a block-shaped memory; wherein, each register block belongs to one of the queues, the block-shape memory is provided with a read-write mode state switching module, both the lock-shaped memory and the read-write mode state switching module are in support of three operational modes and uses the block RAM in the FPGA to store and process all registers in a large number of queues (like 2048 queues), thus realizing simultaneous reading and writing on the related registers in the same queue. The method provided in the invention solves the problems of insufficient logical resources, slow storage speed, long storage cycle and the like in the logical storage and external storage in the prior art.

The invention discloses a DDR management and control system based on FPGA hardware acceleration, the system including a host, an FPGA terminal and a memory DDR. The system is characterized in that theFPGA terminal comprises a DMA transfer control logic unit, a DDR conversion controller and a transmission state finding unit which are connected through an AXI bus, the FPGA terminal also comprises an algorithm logic unit connected to the DDR conversion controller, the FPGA terminal is connected with the host through the DMA transmission control logic unit, the FPGA terminal is connected with thememory DDR through the AXI bus, and the DDR conversion controller is connected with the transmission state finding unit. The system adopts the data transmission state index value in the data processing process and realizes the corresponding data processing operation according to the recorded index value, which simplifies the operation mode of the memory DDR and improves the data reading and writing efficiency by cooperating with the AXI bus. A DMA transmission mechanism is adopted to speed up the data transmission efficiency, reduce the CPU occupancy of the host terminal, and improve the dataprocessing efficiency.

The invention provides a map data processing method and device. The method includes: acquiring information of roads and information of parcels in which the roads are located; according to the information of the roads and the information of the parcels, judging whether the roads belong to different parcels; acquiring first intersection points of the roads and boundary lines among the different parcels if the roads belong to the different parcels; calculating target straight-lines which pass through the first intersection points and are perpendicular to the roads; and using the target straight-lines to divide the roads. According to the scheme, the perpendicular lines are used to divide road attachment objects, which respectively belong to the different parcels, to acquire identification data which have independence and use end points as road lines and attachment object lines, coupling with other data in a map is reduced, a redundancy degree of map data is decreased, and reading / writingefficiency of the map data is improved; and the road lines and the attachment object lines obtained by splitting of the scheme have the same length, and appearing of phenomena of data errors is avoided.

The invention provides a data reading and writing method, a data reading and writing device and a circle queue. The method comprises the steps of acquiring a deviation position of a writing pointer from a queue head of the circle queue; determining a deviation position of a reading pointer according to the deviation position of the writing pointer; and reading data from the circle queue accordingto the deviation position of the reading pointer. According to the data reading and writing method, the data reading and writing device and the circle queue, single-input multi-output of a shared memory is implemented so that multiple reading threads can read data from the circle queue in parallel, thus the data reading and writing efficiency is effectively improved and the memory consumption is reduced.

The invention discloses a communication method of a symbiotic virtual machine based on a multi-core lockless ring buffer area. The communication method comprises the following implementing steps of applying for a buffer area with fixed size in a share memory in advance, equally dividing the buffer area into element data areas, designating one element data area to store description information of the buffer area and identification variables of the buffer area, and organizing the other element data areas to form a ring in logic, so as to form the multi-core lockless ring buffer area; when the symbiotic virtual machine requires communication, using a communication sender as a writer, writing the communication data into the symbiotic virtual machine, and only allowing one writer to execute writing operation at the same time; using a communication receiver as a reader, enabling the reader to read the communication data in the multi-core lockless ring buffer area, and enabling multiple readers to execute the reading operation in a concurrent way. The communication method has the advantages that the high-efficiency concurrent reading by the receivers is supported, the reading and writing concurrency of the buffer area is good, the communication efficiency of the symbiotic virtual machine is high, and the multi-core processing property is fully utilized.