40results about How to "Reduce I/O" patented technology

An electronic device according to various embodiments of the present invention may comprise: a first memory for storing file data; a second memory having an attribute different from an attribute of the first memory; and a processor, wherein the processor is configured to store, in the second memory, journal data associated with a change in at least a part of the file data, and update the at least a part of the file data stored in the first memory at least on the basis of the journal data when a designated condition is satisfied. Other various embodiments are possible.

Apparatus, method and systems for managing reference data, which can prevent duplicated data loading of reference data and eliminate redundancy of I / O operations for loading of the same reference data required by different virtual machines present in the same physical node to reduce use memory and I / O through sharing virtual machine leveled memories, are provided.

The invention discloses a distributed buffer management method based on a dynamic index, applied in a database cluster system, wherein each node of the database cluster system is provided with an independent buffer; when a data block is firstly loaded to the buffer of the node by the certain node, information of the data block is firstly loaded in a dynamic index table of a global buffer manager, the attribute of the buffer is set to be a local resource, and is held by the node; when the other nodes need to access the data block, the nodes firstly search in the global buffer manager, and under the condition that the data block is loaded and held by the certain node is acquired through the dynamic index table, the nodes modify the attribute of the buffer to be a global resource, and apply to the node which holds the data block for transferring the data block for accessing. The invention realizes multi-version assess control of the data block through a dynamic index mechanism, reduces disk I / O (Input / Output), and improves overall performance of the database cluster system.

The invention discloses a quick single-disk failure recovery method for triple-erasure-correcting codes. The method includes: dividing each stripe unit of each stripe into p-1 sub-blocks during recovery of single stripes; acquiring a total number R<idea1> of the sub-blocks, used for recovering a failed disk, in each stripe; in an array of p+2 disks, initializing a feasible solution value for failure of the failed disk I into 3<p-1>; initializing a counter j=0; converting j into a recovered sequence which is ternary and which shares p-1 elements by means of the division algorithm; calculating a sum of bits with 0, 1 and 2 values in the recovered sequence of p-1 bits; comparing the three numbers pairwise; judging whether or not absolutes of differences are all smaller than or equal to 1; if yes, starting an effectiveness test function and judging whether or not test blocks, represented with the recovered sequence, are able to recover all data blocks in the failed disk i. The letter p is a prime number, with p+2 being equal to the number of disks in an RAID (redundant array of independent disks), and R<idea1>=R<low>+w%. The quick single-disk failure recovery method for triple-erasure-correcting codes has the advantages that disk recovery process can be quickened and system reliability is improved.

The present invention provides a data storage method, device and storage engine based on the LSM Tree structure, wherein the method includes processing the data to be stored in the computer memory, parsing out the key of the data to be stored and the value corresponding to the key . Store the value of the data to be stored, and obtain the storage address of the value of the data to be stored. Assign the obtained storage address to the key of the data to be stored, and write the assigned key into the LSM Tree structure. Therefore, by separately storing the key of the data to be stored and its corresponding value value, the LSM Tree structure does not directly store the value value of the data, but stores the key and the corresponding value value storage address for subsequent data processing. Merging can not only effectively reduce a large number of value processing processes, but also reduce disk IO.

The invention discloses a file version control system, which includes a local library and a control module. The local library includes a work area, a metadata database, and a version library. The work area is used to store files managed by the system. For storing file trees, version trees, and version lists, the version library is used to store version files in the work area; the control module includes an abstract file system interface module, a file operation module, a metadata access module, a recovery manager module, a work Area monitor module; the invention encrypts files and backs them up to the server, which can prevent privacy leaks at the server side; automatically maintains version information of files, making it convenient for users to recover to file history versions after misoperation of files. In addition, the file tree, version tree, and version list are stored in a table, and then the table is converted into a key-value pair and stored in a disk file. The modification and deletion of cells in the table are implemented by writing new values, which avoids random access to the disk. Reduce disk I / O and reduce network overhead.

The embodiment of the present application provides a data processing method and related equipment, which can comprehensively utilize the advantages of solid-state equipment, tile recording disks, and log-structured merge trees to optimize the stored data. The method includes: acquiring target data; determining a target ordered key-value pair table corresponding to the target data; determining a second ordered key-value pair table according to the target ordered key-value pair table; The value pair table determines the first ordered key-value pair table; merges the second ordered key-value pair table into the first ordered key-value pair table; when the merged first ordered key-value pair table When the table meets the preset condition, write the merged first ordered key-value pair table and the third ordered key-value pair table as an array set into the target dynamic area whose size dynamically changes in the tile record disk .

The invention provides a method and system for storing and analyzing measurement data format MDF files. The method includes: collecting and uploading MDF data files; sorting the collected MDF data files; extracting metadata from the sorted MDF data files and storing the extracted metadata; Convert the electrical signal value of each record in the file into a physical value; generate a self-parsed columnar storage format file and store the generated columnar storage format file in a distributed storage system; and use a distributed computing system to analyze all Stored columnar storage format files for calculation.