46results about How to "Reduce the number of IOs" patented technology

The invention discloses a large-scale social network service-oriented graph data storage and query method. A data storage manager stores received graph data in a Key-Value way, wherein the vertex ID (Identity) of the graph data is taken as the Key, a vertex neighbourhood is taken as the Value, and the data of each vertex neighbourhood is stored; and a plurality of edges connected with the vertex neighbourhood are orderly stored in a memory block with a fixed size via a timestamp, a double linked list is formed, and the attribution information and the index information of the vertex are stored in a data structure. When the data storage manager receives an access request that the vertex v is accessed, the data storage manager transmits the vertex v and a k-order neighbourhood of the vertex v to a requester; and the requester caches stored data locally, firstly checks local cache during next query, and sends the access request to the data storage manager if no queried vertexes exist. The graph data storage and query method can meet a scene which is dynamically updated and is suitable for processing data sparseness, and random access.

A method for aggregation optimization of temporal data includes dividing a temporal database file into a data block file and a block index file; scanning the block index file according to the calculated start and end time period, extracting the data blocks that meet the time period condition and other filtering conditions, and sorting according to the offset of each data block in the data block file; scanning the data files inexpensively by sorted data blocks, after the calculation of each data block, gathering the calculation results. The method can be completed by opening a data file for scanning whether it is for reading a single time series data or for aggregating multiple time series data, which greatly improves the overall performance.

The present invention provides a virtual IO inter-chip connection circuit for FPGAs. The circuit comprises a clock module, a sending FPGA end circuit, a transmission module and a receiving FPGA end circuit. The sending FPGA end circuit comprises a data coding module for generating coded data; a first asynchronous FIFO which caches coded data written by the data coding module according to a coding clock and an asynchronous FIFO write protocol; and a sending module which sends data of at least two bits and a differential serial clock read from the first asynchronous FIFO until all the coded data is sent. The receiving FPGA end circuit comprises a receiving module for receiving the data of at least two bits and the differential serial clock until all the coded data is received; a second asynchronous FIFO which caches the data of at least two bits written by the receiving module according to the asynchronous FIFO write protocol and the differential serial clock until all the coded data is written; and a data decoding module which decodes the coded data synchronously. According to the invention, the transmission bandwidth and the transmission rate can be increased arbitrarily within a considerable range while the bit error rate can be reduced greatly.

The invention discloses an image display method and device, a storage medium and an electronic device. The method comprises the following steps: target playing time corresponding to a target operatingposition where target operation is performed on a playing progress bar is acquired from a multimedia resource under the condition of detecting the target operation performed on the playing progress bar of the multimedia resource; a target zone corresponding to the target playing time in a target image is determined, wherein the target image comprises a plurality of zones which correspond to a plurality of playing time of the multimedia resource and comprise the target zone, and the plurality of playing time comprise the target playing time; and the image on the target zone is displayed. The technical problem of relatively low image display efficiency in the related art is solved.

The embodiment of the invention discloses a cache file recovery method and device, electronic equipment and a storage medium. The method comprises the following steps of: when an executable file is loaded into the memory, judging whether the executable file is an important cache file or not according to the attribute information of the executable file, marking the executable file if the executablefile is the important cache file, judging whether the residual capacity of the memory meets a preset condition or not, and recycling the unmarked cache file in the memory if the residual capacity ofthe memory does not meet the preset condition. When a user uses the electronic equipment, important cache files and high-frequency access files can be generated in a memory of the electronic equipment, and the unimportant cache files and the non-high-frequency access files are recovered by judging the residual capacity memory of the memory, so that the important cache files are reserved, and the access speed of the files is increased.

The invention discloses an FPGA intra-IO sheet interconnecting digital circuit based on re-encryption algorithm. The FPGA intra-IO sheet interconnecting digital circuit comprises a sending end digital circuit, a receiving end digital circuit and a clock generation module, wherein the sending end digital circuit comprises a plaintext data generation logic, a first asynchronous FIFO, a head synchronous LFSR encryption module, a second asynchronous FIFO, an AES encryption module, a parallel-serial conversion module, a serializer and an OBUFDS module, and the receiving end digital circuit comprises a deseriallizer, an IBUFGDS module, a serial-parallel conversion module, a third asynchronous FIFO write control logic, a third asynchronous FIFO, an AES decryption module, a fourth asynchronous FIFO, a head synchronous LFSR decryption module and a plaintext data reception logic. Encryption and decryption of interconnected data transmission processes between FPGA virtual IO sheets are realized through stable and effective re-encryption algorithm.

The invention discloses a cache writing method. The method comprises the following steps: according to the file index arrangement order in the LRU table, the cached target strip file is read from theSSD; judging whether the number of file fragments of the target strip file is greater than a preset threshold value; if so, the target strip file is aggregated to obtain an aggregated data block; write the aggregated block of data to the HDD via an IO at a time and remove the target stripe file from the SSD. Compared with the mode that each file segment is written into the HDD by IO once at present, the method provided by the embodiment of the invention can make full use of the fast reading and writing performance of the SSD, can improve the writing performance of the storage system and improve the user experience. The invention also discloses a cache writing device, a device and a readable storage medium, which have corresponding technical effects.

The invention relates to the technical field of blockchains, and discloses a multi-version data storage method and device, computer equipment and a storage medium. The method comprises the steps of receiving a modification instruction for specified node data of an MVB tree; copying the first tree node data on the specified node data index path to obtain copied second tree node data; and modifyingthe second tree node data according to the modification instruction, and generating a modified version of the MVB tree according to the node data not in the index path and the modified second tree node data. According to the invention, the storage efficiency of the blockchain data can be improved.

The invention provides an alarm information association method and device, electronic equipment and a computer readable storage medium, and relates to the technical field of network monitoring. The alarm information association method comprises the following steps: when alarm information is collected, carrying out vectorization processing on the alarm information to obtain an alarm vector; obtaining an alarm association model matched with the alarm vector from the alarm association probability model set based on the alarm event represented by the alarm vector; acquiring a plurality of pieces of alarm association information of the alarm vector, and establishing an association relationship between the alarm vector and the plurality of pieces of alarm association information according to the alarm association model; generating an alarm association tree according to the association relationship, and pushing the alarm association tree to the monitoring terminal. According to the technical scheme of the invention, the alarm association tree can be generated in real time based on the alarm information and the alarm association model collected in real time, and the alarm monitoring experience and the alarm response timeliness can be improved.

The invention relates to the field of block chains, and discloses a data storage method and device, computer equipment and a storage medium.The method comprises the steps of newly establishing a transaction, wherein the transaction comprises a plurality of adding instructions; before the transaction is finished, adding a newly-added temporary node according to the adding instruction; after the transaction is finished, splitting the newly added temporary node into a plurality of split nodes, and calculating hash values of the split nodes; updating the key value pair list of the target father node, and judging whether the length of the key value pair list of the target father node exceeds a second preset threshold value or not; and if yes, continuing to split the target father node until thekey value pair list lengths of all nodes do not exceed respective corresponding preset thresholds. Through asynchronous processing of transaction processing and node splitting and coupling of node splitting and hash calculation, the transaction processing efficiency can be improved, the IO frequency of paths and index values of a transaction type Merkel B + tree is reduced, and the disk read-write time is shortened.

The invention discloses a system and method for fast detaching a virtualized disk image. The system comprises a partition table analysis module, a disk analysis module, a file distribution table extraction module, a partition analysis module, a conflict resolution module and a file distribution table write-in module. The partition table analysis module is used for analyzing a target partition and a virtualized disk file physical address. The disk analysis module is used for judging whether a virtualized disk file is within the range of the target partition or not based on the physical address, if not, an error is reported, quitting is carried out, and if yes, the virtualized disk file is analyzed, and a file partition table and a file distribution table are acquired; the file distribution table extraction file is used for judging whether a file in the file partition table is stored in a virtualized disk as it is or not based on the file distribution table and recording file distribution records of files capable of being directly extracted; the partition analysis module is used for analyzing the target partition to acquire the position of a target partition file distribution table; the conflict resolution module is used for comparing the file distribution records with file names in the target partition, and if the file names are identical, file distribution record file names are modified; the file distribution table write-in module is used for writing the file distribution records in the target partition based on the target partition file distribution table and deleting original files.

The invention discloses an access request processing method and device. The method comprises the steps of receiving a plurality of access requests, wherein the access requests are used for accessing astorage bucket; sequentially storing access logs generated by each access request into a cache region, and carrying out merging processing; and writing the current merging result in the cache regioninto the corresponding log storage region according to a preset storage mode in the process of storing the generated access log into the cache region. The technical problem of performance reduction caused by consumption of a large amount of disk IO performance when access requests are frequent in a processing method of the access requests in related technologies is solved.

The invention relates to the technical field of artificial intelligence, and discloses a thread pool management method, device, equipment and medium suitable for a distributed technology, and the method comprises the steps: calling a target application, determining to-be-cached thread pool configuration data from Spring at a preset sleep time interval, and updating a target local cache, acquiring target thread pool working data according to a preset monitoring time interval, updating the server, determining target thread pool configuration data from a target local cache, and determining a queue early warning judgment result according to the target thread pool working data and the target thread pool configuration data, when the queue early warning judgment result is exceeded, generating alarm information to be stored according to the queue early warning judgment result and issuing the alarm information to the server; calling the client to consume the alarm information to be stored in the server; and calling an early warning platform to determine a thread pool statistical result according to the target thread pool working data of the server. Unified management and monitoring of the custom thread pool in the application are realized.

The invention discloses a system and method for fast detaching a virtualized disk image. The system comprises a partition table analysis module, a disk analysis module, a file distribution table extraction module, a partition analysis module, a conflict resolution module and a file distribution table write-in module. The partition table analysis module is used for analyzing a target partition and a virtualized disk file physical address. The disk analysis module is used for judging whether a virtualized disk file is within the range of the target partition or not based on the physical address, if not, an error is reported, quitting is carried out, and if yes, the virtualized disk file is analyzed, and a file partition table and a file distribution table are acquired; the file distribution table extraction file is used for judging whether a file in the file partition table is stored in a virtualized disk as it is or not based on the file distribution table and recording file distribution records of files capable of being directly extracted; the partition analysis module is used for analyzing the target partition to acquire the position of a target partition file distribution table; the conflict resolution module is used for comparing the file distribution records with file names in the target partition, and if the file names are identical, file distribution record file names are modified; the file distribution table write-in module is used for writing the file distribution records in the target partition based on the target partition file distribution table and deleting original files.

A method and system for maintaining tree-structure based directory relationships is provided. The system includes: a change detection unit, used to receive a change request from outside or inside, and set a change flag and record change information related to the change request according to the change request; a polling task unit, used to check polling change flag; query and tree assembly unit, used to assemble tree structure after deduplicating the change information according to the change information; and calculation unit, used to generate according to the tree structure assembled by the query and tree assembly unit directory tree.

The invention provides a bitmap storage method. The bitmap storage method includes the steps that stripes of a redundant array of inexpensive disks raid are evenly distributed according to the number of CPUs, and multiple blocks corresponding to the CPUs one to one are acquired, wherein each block comprises multiple raid stripes; multiple bit pairs are selected from a memory and used for representing the use states of the blocks, and a region bitmap is acquired, wherein the bit pairs comprise first bits and second bits, the first bits are used for representing whether the blocks are written or not, and the second bits are used for representing whether the blocks are initialized. By the adoption of the bitmap storage method, the IO number of the stripe blocks to which a computer has access for the first time after the computer is restarted can be reduced, the first-time IO write speed is higher after the computer is restarted, when the computer is restarted, the initialization and reconstruction progress is kept to a large degree, and multiple restarts are allowed in the reconstruction or initialization process; the storage region of the bitmap is saved, and then the probability that the bitmap in superblocks breaks down is reduced.

The embodiment of the invention provides a data retrieval method and a server. The method comprises the steps: obtaining a core data table in data tables in a database; parsing a query sentence of thecore data table; obtaining a grouping reference column according to a first preset rule and the parsing result; and according to a second preset rule and the grouping reference column, obtaining an index constructed based on the core data table, and performing data retrieval according to the index. The server performs the method described above. The data retrieval method and the server provided by the embodiment of the invention can greatly reduce the IO quantity of the database and the storage space, thereby remarkably improving the efficiency of the overall operation of the system.

The embodiment of the invention provides a big data write-in method and device, a big data read method and device and a big data processing system. The write-in method includes the steps that a write-in request is received; big data are segmented into multiple pieces of subdata according to the write-in request; the subdata are uniform in length, and relative positions of data parts, corresponding to the same field, in the subdata are fixed; the subdata are written in a specified storage medium; storage positions, in the storage medium, of current subdata are added to all the subdata. The big data can be positioned easily, conveniently and rapidly, time consumption is reduced, the speed of storing and reading the big data is increased, and system performance and working efficiency are improved.

The invention discloses a large-scale social network-oriented graph data storage and query method. The data storage manager of the invention stores the received graph data in a Key-Value manner, uses the vertex ID of the graph data as the Key, and uses the vertex neighbor The field is Value; data storage for each vertex neighborhood: store multiple edges connected to the vertex neighborhood in a fixed-size memory block in order with time stamps, and form a doubly linked list, and attribute information of the vertex and index information are stored in a data structure. When the data storage manager receives an access request to visit vertex v, the data storage manager transmits the vertex v and its k-order neighbors to the requester; the requester caches the returned data locally, and the next query, first check The local cache, if there is no queried vertex, sends the access request to the data storage manager. The invention can satisfy dynamic update, and is suitable for processing sparse data scenarios and random access.