25results about How to "Improve read performance" patented technology

The invention provides a solid state disk channel signal adjusting system and a solid state disk, the system further comprises a main control chip, a memory chip, a real-time monitoring module and a time sequence adjusting module, the real-time monitoring module is used for monitoring a read operation error rate corresponding to a flash memory chip in real time, and when the read operation error rate is larger than a preset threshold value, the time sequence adjusting module is used for adjusting the time sequence of the flash memory chip; working environment parameters of the DQS signal and the DQ signal are obtained and transmitted to the main control chip; the main control chip is used for obtaining a time sequence adjusting signal of a DQS signal and a DQ signal corresponding to the flash memory chip according to the working environment parameters and transmitting the time sequence adjusting signal to the time sequence adjusting module; wherein the time sequence adjusting signal is used for realizing alignment of a DQS signal and a DQ signal corresponding to the flash memory chip; and the time sequence adjusting module adjusts the time sequence of the DQS signal and the DQ signal corresponding to the flash memory chip according to the time sequence adjusting signal. According to the solid state disk channel signal adjusting system and the solid state disk, self-adaptive optimization adjustment of the solid state disk channel signals is achieved, the solid state disk channel signal adjusting system and the solid state disk can adapt to complex and changeable working environments, and the reading performance and the data transmission rate of the solid state disk are remarkably improved.

The invention relates to a semiconductor structure and a manufacturing method thereof. The semiconductor structure comprises a substrate; the memory unit is arranged on the substrate, the memory unit comprises a transistor and a capacitor, the transistor comprises a first source / drain electrode, a grid electrode, a second source / drain electrode and a semiconductor layer surrounding the side wall of the grid electrode, the first source / drain electrode, the second source / drain electrode and the semiconductor layer are sequentially arranged in the first direction, and the first source / drain electrode and the second source / drain electrode are connected with the grid electrode through the semiconductor layer; the capacitor is arranged on one side of the second source / drain electrode and covers part of the side surface of the semiconductor layer; the bit line is arranged on one side of the first source / drain electrode and covers part of the side face of the semiconductor layer, the contact area of the bit line and the semiconductor layer is smaller than the contact area of the capacitor and the semiconductor layer, the contact area of the bit line and the transistor is reduced, and the opposite area of the bit line and the grid electrode of the transistor is reduced; therefore, the parasitic capacitance between the bit line and the grid electrode of the transistor is reduced, and the reading capability of the transistor is improved.

The invention relates to the technical field of computer networks, in particular to an implementation method and system for an RDMA network card request queue, and the method comprises the steps: checking whether a residual space exists in an RDMA Cache module or not when a write request is received; if yes, writing the writing information of the writing request into an RDMA Cache module, updating a free bit, and recording a queue pair number and an unfinished pointer; if not, writing the writing information of the writing request into a host memory module; when a read request is received, checking whether the queue pair number and the unfinished pointer carried by the read request are matched with historical records in the RDMA Cache module or not; if yes, reading from the RDMA Cache module; and if not, reading from the host memory module. According to the invention, multiple QPs in the RDMA network card share the internal resources of the Cache, the resource consumption is reduced, and the utilization rate of the cache is improved.

The invention discloses an analysis query method and device for a database, and the method comprises the steps: obtaining data corresponding to a query instruction from a pre-constructed preset storage structure through a preset index algorithm when the query instruction is received in a data query process of a database system, the preset storage structure is obtained by optimizing a storage structure design layer of a multi-version concurrency control mechanism, the preset storage structure is used for avoiding volume expansion caused by multi-version control and improving the performance of data analysis and query, and data is compressed through a data compression algorithm. And storing the compressed data in a preset storage structure in a column storage mode.

The present invention provides a storage device and a data reading method thereof, wherein the storage device comprises: a flash memory particle, provided with a page register, the page register is used to temporarily store host data in a storage page of the flash memory particle; a main controller, provided with a read cache area, the read cache area is used to temporarily store part of the host data in the page register; wherein the main controller is used to respond to a read instruction of the host to obtain a logical address segment; the main controller is also used to sequentially query the mapping relationship of the logical address segment to obtain the host data in the storage page corresponding to the starting logical block address in the query, and write the host data into the page register; the main controller is also used to write part of the physical block address mapped by the logical address segment in the page register into the read cache area; the main controller is also used to transmit the data in the read cache area to the host when the read cache area is full. The present invention can improve the data processing speed and enhance the reading performance of the storage device.

The invention relates to a data reading method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: receiving a data reading request, wherein the data reading request carries a target key; in response to the data reading request, whether a target Key exists in a static cache region or not is inquired, and the number of historical access times of active key value pairs included in the static cache region is larger than a preset threshold value; and under the condition that the target Key exists in the static cache region, obtaining a target value Value corresponding to the target Key from the active key value pairs included in the static cache region. By adopting the method, the reading performance can be improved.

The application provides a data caching method and device, equipment, medium and product, and relates to the technical fields of big data and artificial intelligence. The method comprises the following steps: obtaining a early warning task by a master node in a cluster, the early warning task being used for calculating data stored in a data table according to a pre-designed calculation rule to obtain an early warning cache value, the cluster comprising the master node and a slave node; performing calculation according to the early warning task to update the early warning cache value in the early warning task; and synchronizing the updated early warning task to the slave node, the slave node being used for receiving a reading request and returning the early warning cache value obtained from the early warning task of the slave node. When a user needs to obtain the early warning cache value, the user can send a reading request to the slave node, and the slave node returns the early warning cache value to the user. Since the early warning cache value is obtained in advance, the same reading request of the user can not be calculated every time, and frequent reading of data in the data table is avoided.

The invention discloses an RFID jewelry label reading system and method, and relates to the technical field of jewelry label reading, a plurality of reading schemes are generated based on a database, after Monte Carlo simulation is carried out on each reading scheme, an expression value of each reading scheme is calculated by using an expression function, and if there is a reading scheme with an expression value greater than an expression threshold, the reading scheme is read. If yes, the reading scheme with the maximum performance value is selected as the final scheme of the reading system for use, otherwise, part of the reading schemes are selected for content optimization iteration, iteration is stopped until the reading scheme with the performance value larger than the performance threshold value exists, and the final scheme is output for the reading system for use. Monte Carlo simulation is introduced into the reading system, performance values of different reading schemes are evaluated through randomization experiments, after the schemes can be comprehensively measured, the optimal scheme is selected for the reading system to use, the reading performance is gradually improved, and high efficiency and reliability of the final scheme are ensured.

This invention relates to the field of storage system IO performance optimization technology, and provides a method and device for redirecting data aggregation, comprising: identifying random small IO data from written IO requests, aggregating and writing to backend storage objects, and constructing a storage mapping relationship in an index table; traversing backend storage objects storing random small IO data, and allocating data originating from the same logical address space to the same target backend storage object using a hash algorithm; sequentially writing the data allocated to the same target backend storage object into the contiguous physical space of the target backend storage object, and updating the mapping relationship in the index table; repeating steps S2 and S3 before reaching a preset termination trigger condition, using a different hash algorithm each time step S2 is repeated; when reading data, retrieving the storage mapping relationship from the index table, and aggregating the read requests corresponding to data located in the same backend storage object before reading them all at once.

The invention discloses a block chain-based data processing method, apparatus and device, and a readable storage medium. The method comprises the steps of obtaining first transaction data and a transaction type to which the first transaction data belongs; according to the transaction type, determining a target fragment used for caching the first transaction data in N fragments set in a transaction pool of the block chain; the block chain comprises N sub-chains, and the N sub-chains share a transaction pool of the block chain; the N fragments are divided in a transaction pool of the block chain based on N sub-chains, and one fragment corresponds to one sub-chain; and acquiring a node partition corresponding to the first consensus node from M node partitions arranged in the target partition, and caching the first transaction data to the node partition corresponding to the first consensus node. By adopting the method and the device, the cache redundancy of the transaction data can be reduced in the transaction pool cache business, and meanwhile, the reading performance of the transaction data is optimized.

A reading device for obtaining information of a specified product, the reading device of the embodiment includes a housing, an imaging section, and an illumination section. The housing includes a transparent reading window. The imaging section is built into the housing and images an image across the reading window. The illumination section illuminates an imaging object of the imaging section by surface emission toward the outside of the housing, is disposed outside the edge of the reading window, and is inclined to the reading window in a manner that protrudes from the reading window more on a side farther from the reading window.

The invention provides a table reading method and device in an actuarial model, computer equipment and a medium. The method comprises the following steps: analyzing variables to be used in the calculation process of an actuarial model to obtain target variables; obtaining calculation parameters for limiting the target variable according to the calculation task; determining a table where the target variable is located as a target table; querying in the target table according to the calculation parameters; judging whether the currently queried data is a table name or a constant; if the currently queried data is the table name, continuing to query in the table corresponding to the table name so as to realize nested table query; and if the currently queried data is the constant, determining the value of the target variable according to the constant. Through the method, the data volume can be reduced when the table is read in the actuarial model, the table data reference logic understanding difficulty is reduced, and the table data reading speed is increased.

The application discloses a hard disk garbage collection method and device. The method comprises the following steps: dividing a storage object to be collected into a plurality of fixed-size continuous intervals, determining the starting and ending intervals of the data to be migrated according to the offset and length of the data to be migrated; dividing a reading scene into three categories, i.e., a single-interval, a cross-double-interval and a cross-multiple-interval, based on the interval position relationship; in the single-interval scene, if the interval has been cached, the data is directly copied, otherwise, the complete interval data is read to the cache when the reading IOPS does not reach the upper limit; the same logic is executed for each interval in the cross-double-interval scene; in the cross-multiple-interval scene, the data is directly read and the IOPS concurrency is converted according to the actual length; and the effective data is migrated to a new storage object and the original object space is released. The application effectively reduces the number of hard disk reading requests, reduces the mechanical disk head addressing overhead, simultaneously reduces the IOPS flow control conflict by using cache hits, and significantly improves the garbage collection efficiency of the storage system and the stability of the cluster capacity.

The present invention discloses a skip read pre-reading method, device, equipment and storage medium, the method comprising: receiving a read request, if the data corresponding to the read request is already stored locally, then reading the data corresponding to the read request locally, otherwise, reading the data corresponding to the read request from a target file; after reading the data corresponding to the read request, determining whether the current read mode is a skip read mode, if so, predicting the pre-read data position and pre-read data length of the data to be read after the current moment based on the skip read mode, pre-reading the data corresponding to the pre-read data position and the pre-read data length from the target file, and storing the pre-read data locally, otherwise, determining that no pre-reading operation is required; the pre-reading operation is an operation of pre-reading corresponding data from the target file. Therefore, in the scenario where the read operation is implemented in the skip read mode, the present application can improve the data reading speed and the reading performance to a certain extent by pre-reading data.

The application belongs to the technical field of solid-state storage, and specifically discloses a solid-state disk fault-aware reading optimization and code length self-adaption method and device, which comprises the following steps: a fault-aware re-reading mechanism is constructed, a code word that fails to be decoded in a primary decoding round is marked, and only the marked failed code word is transmitted in a subsequent re-reading round; based on the fault-aware re-reading mechanism, the delay of hard-decision decoding and the delay of soft-decision decoding are analyzed and combined to obtain a fault-aware reading delay model; the fault-aware reading delay model comprises a default reading operation delay model and a soft-decision reading operation delay model; when reading data, a reading mode determination result is obtained based on the fault-aware reading delay model; when writing data, a target code length in the case of the shortest reading delay is obtained based on the fault-aware reading delay model; and the target code length is used to write data. Through the application, the reading performance of an SSD can be improved under the condition that the code rate remains unchanged.

The present invention relates to a controllable real-time task remote execution system, comprising a task management module, a task scheduling module and a task execution backend; the task management module is connected to the task scheduling module, and the task management module is used to create a task and send it to the task scheduling module; the task scheduling module is connected to the task execution backend, and the task execution backend is used to obtain and execute a task scheduling object; the task execution backend is connected to the task management module TCP long connection, and the task execution backend is used to transmit the task scheduling object execution real-time status and / or real-time log information to the task management module. Through the task management unit and the task execution backend TCP long connection, the information synchronization efficiency is improved, which facilitates the task management unit to intervene in the task execution backend in time when the task execution is abnormal; and solves the problem that the information synchronization method in the prior art adopts a polling mechanism, causing information delay and execution system performance loss.

The invention provides a monitor central station ordered data storage method, an indexing method and computer equipment, and the method comprises the following steps: an ordered data storage step: receiving real-time data reported by a monitor, extracting the ordered data, converting the ordered data into a file storage data structure, compressing the data structure, and adding the compressed data structure to a latest ordered data file; a global offset obtaining step: after the ordered data is written, calculating the global offset of the ordered data by using the offset written into the ordered data file; and an index data storage step: constructing index information according to the global offset, calculating the position of the current data in the index data file by using the data timestamp, and writing or updating the index data to the corresponding position. According to the method, the classification and segmentation storage method is optimized on the basis of monitoring vital sign data characteristics, the monitoring data storage duration is prolonged, efficient, continuous and ordered indexes are established for waveform data and measured value data on the basis of the time sequence, the review data reading performance can be greatly improved, and efficient support is provided for other analysis and viewing.

The invention relates to the technical field of data storage, and provides a cold and hot hierarchical storage and SLC bypass management method based on a storage class memory SSD and a ZNS SSD, which comprises the following steps: step 1, system initialization and heterogeneous hierarchical mapping; 2, request receiving and read-write shunting are carried out; when an application layer initiates a key value operation request, the system firstly judges the type of the request: if the request is a write request, entering step 3; if yes, entering step 7; step 3, writing and solidifying data; 4, temperature judgment and SLC bypass judgment are carried out; step 5, overlap detection and target level judgment; step 6, executing heterogeneous medium writing; step 7, when a read request is processed, performing cascading search according to medium performance from high level to low level, and realizing read optimization of physical isolation; and step 8, ending: returning an operation result to the application layer. According to the invention, cold and hot hierarchical storage and SLC bypass management can be well carried out.

The present invention discloses an RFID jewelry tag reading system and method, which relates to the field of jewelry tag reading technology. A plurality of reading schemes are generated based on a database. After Monte Carlo simulation is performed on each reading scheme, a performance function is used to calculate the performance value of each reading scheme. If there is a reading scheme with a performance value greater than a performance threshold, the reading scheme with the largest performance value is selected as the final scheme of the reading system. Otherwise, some reading schemes are selected for content optimization iteration until there is a reading scheme with a performance value greater than the performance threshold, and the iteration is stopped, and the final scheme is output for use by the reading system. The reading system introduces Monte Carlo simulation, and evaluates the performance values ​​of different reading schemes through randomized experiments. After comprehensively measuring the schemes, the optimal scheme can be selected for use by the reading system, and the reading performance can be gradually improved to ensure the efficiency and reliability of the final scheme.

The disclosure relates to a storage device for storing an artificial intelligence model and an operating method thereof. The storage device can include a memory configured to store a plurality of data units each including at least one of parameters of an artificial intelligence model, and a controller configured to receive structure information about the artificial intelligence model from the outside, determine the plurality of data units based on the structure information, and determine whether to cache each of the plurality of data units in a cache.

The embodiment of the invention provides a memory device and an operation method thereof, a memory system and a storage medium. The memory device comprises a memory cell array and a peripheral circuit. The memory cell array comprises a plurality of memory blocks; each storage block comprises a plurality of storage units and a plurality of word lines; the peripheral circuit is configured to apply a first pass voltage on the first word line, a second pass voltage on the second word line, and a third pass voltage on the third word line when a first read operation is performed on a memory cell coupled to a selected word line; the first word lines are at least one word line which is physically located above and below the selected word line respectively; the second word line and the third word line are both word lines which are physically located on the side, away from the selected word line, of the first word line, the storage unit coupled to the second word line is a programmed storage unit, and the storage unit coupled to the third word line is an unprogrammed storage unit; the first passing voltage, the second passing voltage and the third passing voltage are different.

The embodiment of the invention provides a method and device for determining a logic block address access sequence and a program product, and the method comprises the steps: collecting an access sequence set of logic block addresses from solid-state storage equipment; determining M target logic block areas from the N logic block areas by utilizing the access sequence set; a logic block adjacency matrix of the M target logic block areas is constructed, and the logic block adjacency matrix is used for representing the transfer frequency between the M target logic block areas; and determining H target logic block access sequences based on the logic block adjacency matrix. Through the method and the device, the problem that the training complexity of the pre-reading model is increased due to sparse distribution and poor regularity of the LBA sequence data in related technologies is solved, and the effects of effectively improving the reading performance of the solid-state storage equipment and enabling the pre-reading model to more accurately predict and pre-fetch the data are achieved.

The utility model discloses a cable head temperature monitoring system, which comprises a central processing unit, a temperature sensor, a temperature collector, a radio frequency gain antenna and a power management module, and is characterized in that the central processing unit is respectively connected with the temperature sensor, the temperature collector, the radio frequency gain antenna and the power management module; and the radio frequency gain antenna is respectively connected with the temperature sensor and the temperature collector. According to the cable head temperature monitoring system, the temperature collector, various temperature sensors and the radio frequency gain antenna are matched, so that the whole temperature monitoring process can be completed synchronously, and the safety, the convenience and the real-time performance of the cable head temperature monitoring system are improved.

The application relates to the technical field of data processing, and discloses a data writing method, a flash memory device and a computer readable storage medium. The method comprises the following steps: acquiring file block data brushed by a host, wherein one file block data corresponds to one file block; identifying the file block data based on a machine learning model to determine the data type of each file block, wherein the data type of the file block is several types; determining several file blocks of the same data type according to the data type of each file block; and sequentially writing the several file blocks of the same data type into several dies, wherein the file blocks of the same data type and the dies sequentially correspond. The data type of each file block is identified through the machine learning model, and the file blocks of the same type are sequentially and cyclically distributed on the dies. The application can reduce die-level conflicts and improve the reading performance of the flash memory device.

The invention discloses a method and system for reducing storage synchronization delay of a cloud edge object storage system, and belongs to the field of object storage, and the method comprises the following steps: abstracting a newly-added SSD supporting a ZNS standard and an original cloud edge object storage system into an integral system, and receiving a request from an upper layer; when a synchronous file adding request is received, the sum of the file size and the size of the used space of the added SSD and the utilization rate eta of the original cloud edge object storage system are calculated, and if any one of (a)-(c) is met, a memory used for storing the file is determined to be the added SSD; the method comprises the following steps: (a) a sum size; the method comprises the following steps of: performing defsize; (b) Desize is smaller than or equal to sum size; maxsize and [eta] gt; hdduti is carried out; (c), max size is smaller than or equal to sum size; ssdsize and [eta] gt; hdduti is carried out; and when (c) is satisfied, the newly-added SSD does not accept the write request any more until the used space is smaller than the maxsize. According to the method, the storage architecture and the request processing mode of the cloud edge object storage system can be improved, and the file synchronization delay is reduced.