CN105630413A - Synchronized writeback method for disk data - Google Patents
Synchronized writeback method for disk data Download PDFInfo
- Publication number
- CN105630413A CN105630413A CN201510976637.9A CN201510976637A CN105630413A CN 105630413 A CN105630413 A CN 105630413A CN 201510976637 A CN201510976637 A CN 201510976637A CN 105630413 A CN105630413 A CN 105630413A
- Authority
- CN
- China
- Prior art keywords
- data block
- write
- data
- block
- diff
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/061—Improving I/O performance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0638—Organizing or formatting or addressing of data
- G06F3/0644—Management of space entities, e.g. partitions, extents, pools
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0655—Vertical data movement, i.e. input-output transfer; data movement between one or more hosts and one or more storage devices
- G06F3/0656—Data buffering arrangements
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Memory System Of A Hierarchy Structure (AREA)
- Signal Processing For Digital Recording And Reproducing (AREA)
Abstract
The present invention discloses a synchronized writeback method for disk data. The method comprises: setting a write cache space, comparing and acquiring difference calculation results diff of data blocks A and A' that have the same start sector number, and substituting the data block A for the data block A'; assigning a weight value to each difference calculation result diff, and sorting the difference calculation results diff from largest to smallest; when a common data block B requests to be written, controlling to perform compression on the data block B, so as to save partial space, taking and inserting the difference calculation result diff with the largest weight into the partial space, and writing a newly-formed data block B into a Z-cluster whose first block is empty of a disk, and caching the data block B into the write cache space; and when a utilization rate of a write cache reaches a preset threshold, substituting a data block in the write cache for the data block on the disk, and changing the Z-cluster into an N cluster. The method improves an I/O bandwidth utilization rate of the disk, and effectively avoids write amplification of small data.
Description
Technical field
The present invention relates to computer data read-write technical field, particularly relate to a kind of write-back I/O dispatching patcher based on synchronous write.
Background technology
The minimum logical addressable unit of shelves system is block (block), and the size of block is generally 4KB. Addressable unit minimum in block device is sector, and the sector-size of most of block devices is 512B, and a block is made up of 8 sectors. Physically, disk is made up of many sectors, and in logic, disk is made up of many block devices.
I/O scheduler is one layer of software between grade system and block storage device. Application program, by calling read/write method or mmap method, creates I/O request, and these requests can be to the amendment of optional position on any shelves. Shelves system will not receive order by request simply and submit request to, without being immediately stored in disk upon receiving a request. But before submission, it can first carry out merging and sorting operation, thus improving the overall performance of system.
At present, existing operating system is write data generally in the following way: when to shelves system writing data blocks, shelves system needs amendment metadata, and these metadata are generally only some bytes, in order to ensure the High Availabitity of system, shelves system needs to store under disk the amendment of these metadata. The read-write of disk is in units of block, usual block be sized to 4KB, for asynchronous write, amendment can be run up to and to a certain degree write disk again by system, and for synchronous write, when writing the small data of several bit groups to shelves system, shelves system must write 4KB data to disk, thus what cause small data writes amplification, waste the I/O bandwidth of disk, the performance of influential system.
Summary of the invention
It is contemplated that one of technical problem solved at least to a certain extent in correlation technique. It is an object of the invention to propose a kind of synchronization write-back method of data in magnetic disk for this.
A kind of synchronization write-back method of data in magnetic disk, described disk logical partitioning is multiple Z-memory block and multiple N-memory block, described each Z-memory block is formed by multiple Z bunch, described N-memory block is formed by multiple N bunch, described Z-bunch and N-bunch is respectively provided with multiple pieces, each piece is made up of multiple sectors, and described synchronization write-back method includes:
Arrange one and write buffer memory, relatively and obtain grade system and need the data block A and the described difference result of calculation diff writing the data block A ' in buffer memory with identical start sector number of write, and the data block A ' write in buffer memory is replaced with data block A;
Adopting the result diff that each difference calculates by absolute fair algorithm to give weighted value, result diff difference calculated according to weighted value is from significantly little sequence;
When there being new general data block B request write, control data block B is compressed, save segment space, the result diff of the difference calculating that weighting weight is maximum is inserted into described segment space and forms data block B ', and to control to be written to data block B ' first block of disk be in empty Z-bunch, and data block B is cached to writes in spatial cache;
When the described utilization rate writing buffer memory reaches predetermined threshold value, displace the data block on the data block Replace Disk and Press Anykey To Reboot write in buffer memory, make Z-bunch to become N bunch.
Preferably, described arrange one and write buffer memory, relatively and obtain grade system and need the concretely comprising the following steps of data block A and the described difference result of calculation diff writing the data block A ' in buffer memory with identical start sector number of write:
Shelves system receives write request, the start sector number of the described data block A in acquisition write request, and searches in writing buffer memory, if there is the data block A ' of identical start sector number;
If it is present use difference to calculate, calculate the difference result of calculation diff between data block A and data block A ';
If it does not exist, then directly write request is submitted in request queue.
Preferably, when there being new general data block B request write, further comprised the steps of: before controlling data block B is compressed
Judge whether variant result of calculation diff;
If variant result of calculation diff, partial block pattern persistent storage data mode is then adopted first data block B to be compressed, save segment space, calculate the size Len of the size data block B after compression, and the result diff that the heavy maximum difference of weighting calculates is inserted in described saved segment space and forms data block B ', and data block B ' is written to described disk;
If zero difference result of calculation diff, adopt full block pattern persistent storage data mode to update disk logical structure, then described data block B is written in the block after first data block of described Z race.
Preferably, the result diff that the difference that described weighting weight is maximum calculates is inserted into concretely comprising the following steps in described saved segment space:
Difference result of calculation diff is compressed, and uses timestamp to identify this difference result of calculation diff, then the difference result of calculation diff with timestamp is embedded in described saved segment space and forms data block B ';
Again write request is submitted in request queue, and data block B is cached to writes in buffer memory, and the read operation service of these part data is provided.
Preferably, described partial block pattern persistent storage data mode is in disk, the data block write in buffer memory providing the read operation of these part data by data block B ' storage with variant result of calculation diff in request queue.
Preferably, described displace the data block on the data block Replace Disk and Press Anykey To Reboot write in buffer memory when the described utilization rate writing buffer memory reaches predetermined threshold value, make Z-bunch to become N bunch concretely comprise the following steps:
If writing utilization rate in buffer memory to have reached threshold value, then use replacement algorithm, select the some data block replacement write in spatial cache to go out, by full block pattern persistent storage data mode, this data block is covered the data block B ' of previous employing partial block pattern perdurable data storage mode storage.
Preferably, described synchronization write-back method also includes:
For the read request that system is submitted to, in writing buffer memory, search whether to meet the data of requirement;
If it has, then directly take out these data to provide reading service;
Otherwise, read request is ranked up, is then distributed in read request queue from high to low by weight, then read data to disk, it is provided that the reading service of shelves system.
Preferably, described synchronization write-back method also includes: after system reboot, and by scanning disk, search embedded in the data block of difference result of calculation diff, and these data blocks is loaded into and writes in spatial cache, provides the read operation of these part data for shelves system.
Preferably, the size respectively 64MB of described Z-memory block and N-memory block, the size of described Z bunch and N bunch respectively 64KB, described piece be sized to 4KB.
Preferably, described arrange one and write concretely comprising the following steps of buffer memory:
One piece of write-back memory space is distributed, for buffer unit divided data block, it is provided that the read operation of these data for I/O scheduler.
The synchronization write-back method of the data in magnetic disk of the embodiment of the present invention, by the data block with identical start sector number is carried out difference calculating, obtain difference result of calculation Diff, cross and difference result of calculation Diff is embedded in general data block, it is achieved synchronous write small data, so that small data will not be lost, a certain amount of disk is write again when small data is cached to, can being effectively improved the write performance of small data, thus improving the utilization rate of magnetic disc i/o, what be prevented effectively from small data writes amplification.
Accompanying drawing explanation
Fig. 1 is the flow chart synchronizing write-back method one embodiment according to inventive disk data;
Fig. 2 is the schematic diagram of the disk logical structure of the present invention;
Fig. 3 is the write-back method that synchronizes according to inventive disk data to the schematic diagram of the write data block with difference result of calculation in Z bunch;
Fig. 4 is the write-back method that synchronizes according to inventive disk data to the schematic diagram of write general data block in Z bunch;
The general data block in write-back method that synchronizes that Fig. 5 is according to inventive disk data covers the schematic diagram of the data block with variant result of calculation.
Detailed description of the invention
Being described below in detail embodiments of the invention, the example of described embodiment is shown in the drawings, and wherein same or similar label represents same or similar element or has the element of same or like function from start to finish. The embodiment described below with reference to accompanying drawing is illustrative of, it is intended to is used for explaining the present invention, and is not considered as limiting the invention.
Below with reference to the accompanying drawings the coded method of the user tag of the embodiment of the present invention and the code device of user tag are described.
As in figure 2 it is shown, disk to be carried out the logical partitioning of more coarseness, build a logical structure imageformat. What disk was divided into some memory block zone, each memory block zone is preferably sized to as 64MB, is divided into Z-memory block (Z-zone) and N-memory block (N-zone) two kinds. Each memory block zone is preferably made up of 1024 bunches of cluster again, each bunch of cluster is preferably sized to 64KB, bunch cluster is divided into Z bunch (Z-cluster) and N bunch (N-cluster) two kinds, and each bunch of cluster is preferably made up of 16 block block.
Z-memory block (Z-zone) can only be made up of Z bunch (Z-cluster), N-memory block (N-zone) can only be made up of N bunch (N-cluster), first piece (block) of Z bunch (Z-cluster) is compressed and be embedded in the result diff that difference calculates, difference calculating is the content by comparing two identical data block block of start sector number, calculates the difference diff of data. The rest block block of Z bunch (Z-cluster) is the block block being not embedded with the result diff that difference calculates normally. N bunch (N-cluster) is made up of the block being not embedded with the result diff that difference calculates normally. Concrete data storage procedure is detailed below.
Fig. 1 is the flow chart synchronizing write-back method according to one embodiment of the invention data in magnetic disk. The synchronization write-back method of data in magnetic disk includes:
Arrange one and write spatial cache, relatively and obtain grade system and need the data block A and the described difference result of calculation diff writing the data block A ' in buffer memory with identical start sector number of write, and the data block A ' write in buffer memory is replaced with data block A;
Adopting the result diff that each difference calculates by absolute fair algorithm to give weighted value, result diff difference calculated according to weighted value is from significantly little sequence;
When there being general data block B request write, control data block B is compressed, save segment space, the result diff of the difference calculating that weighting weight is maximum is inserted into described segment space and forms data block B ', and to control to be written to data block B ' first block of disk be in empty Z-bunch, and data block B is cached to writes in spatial cache;
When the described utilization rate writing buffer memory reaches predetermined threshold value, displace the data block on the data block Replace Disk and Press Anykey To Reboot write in buffer memory, make Z-bunch to become N bunch.
Further, described arrange one and write spatial cache, relatively and obtain grade system and need the data block A and the described difference result of calculation diff writing the data block A ' in buffer memory with identical start sector number of write, and the data block A ' write in buffer memory is replaced with concretely comprising the following steps of data block A:
Distribute one piece of write-back memory space writebackbuffer for I/O scheduler, namely write spatial cache, for buffer unit divided data block block, it is provided that the read operation of these data.
Shelves system receives write request, obtain the relevant information of data block A in write request, such as data block A start sector number on disk, by comparing with the information of the data block in the writebackbuffer of write-back memory space, find out the data block A ' that whether there is identical start sector number;
If it is present use difference to calculate, calculate the difference result of calculation diff between data block A and data block A ', and data block A is covered data block A ';
If it does not exist, then directly write request is submitted in request queue.
Shown in Fig. 3 and Fig. 4, it is preferable that when there being new general data block B request write, further comprised the steps of: before controlling data block B is compressed
Judge whether variant result of calculation diff;
If variant result of calculation diff, partial block pattern persistent storage data mode is then adopted first data block B to be compressed, save segment space, calculate the size Len of the size data block B after compression, and the result diff that the heavy maximum difference of weighting calculates is inserted in described saved segment space and forms data block B ', and data block B ' is written to described disk;
If zero difference result of calculation diff, adopt full block pattern persistent storage data mode to update disk logical structure, then described data block B is written in the block after first data block of described Z race.
The result diff of the difference calculating that weighting described above weight is maximum is inserted into concretely comprising the following steps in described saved segment space:
Difference result of calculation diff is compressed, and uses timestamp to identify this difference result of calculation diff, then the difference result of calculation diff with timestamp is embedded in described saved segment space and forms data block B ';
Again write request is submitted in request queue, and data block B is cached to writes in buffer memory, and the read operation service of these part data is provided.
In the present embodiment, adopt and be similar to the result diff imparting weighted value that each difference is calculated by absolute fair algorithm CFQ, completelyfairqueueing, be ranked up according to order from big to small according to the result diff that difference is calculated by weight.
Preferably, if one first block of existence is empty Z-cluster in logical structure imageformat, then this data block B ' is write this position, if it does not exist, then this data block be written in a new Z-cluster; Need to be written into without difference result of calculation diff, adopt full block pattern full-blockmodel persistence mode to store data, update logical structure imageformat, then this data block B is written in the block after first data block of Z-cluster.
Partial block pattern persistent storage data mode described in above-described embodiment is in disk, the data block write in buffer memory providing the read operation of these part data by data block B ' storage with variant result of calculation diff in request queue.
It is highly preferred that in conjunction with, shown in Fig. 5, when the described utilization rate writing buffer memory reaches predetermined threshold value, displacing the data block on the data block Replace Disk and Press Anykey To Reboot write in buffer memory, Z-bunch is made to become N bunch concretely comprise the following steps:
If write-back memory space utilization rate has reached threshold value, then use replacement algorithm, displace the data block in write-back memory space, and adopt full block pattern full-blockmodel persistence mode to store data, and this data block is covered the previous data block adopting partial block pattern partial-blockmodel persistence mode to store, and update logical structure imageformat, data block owing to being partially submerged into difference result of calculation diff is covered by normal data block, this bunch of cluster becomes N bunch (N-cluster), so also just Z bunch (Z-cluster) embedded in the data block of relevant difference result of calculation diff being become N bunch (N-cluster).
Further, described synchronization write-back method also includes:
For the read request that system is submitted to, in writing buffer memory, search whether to meet the data of requirement;
If it has, then directly take out these data to provide reading service;
Otherwise, read request is ranked up, is then distributed in read request queue from high to low by weight, then read data to disk, it is provided that the reading service of shelves system.
Above technical scheme, in the process implemented, when the machine of delaying occurs in system something unexpected happened, by scanning disk, search embedded in the data block of difference result of calculation diff, and these data blocks are loaded into write in spatial cache, the read operation of these part data is provided for shelves system, returns to the system mode before the machine of delaying.
The synchronization write-back method of the data in magnetic disk described in the above embodiment of the present invention, by using write-back mechanism, accumulation small write operation, when reached a certain amount of after, write disk again, improve the utilization rate of magnetic disc i/o bandwidth, what effectively prevent small data writes amplification, the program is the loss preventing small data simultaneously, by data block is compressed, saves memory space, and the small data after compression is embedded in the space after saving, write disk together, thus reaching the purpose of synchronous write small data.
Although above it has been shown and described that embodiments of the invention, it is understandable that, above-described embodiment is illustrative of, it is impossible to be interpreted as limitation of the present invention, and above-described embodiment can be changed, revises, replace and modification by those of ordinary skill in the art within the scope of the invention.
Claims (10)
1. the synchronization write-back method of a data in magnetic disk, described disk logical partitioning is multiple Z-memory block and multiple N-memory block, described each Z-memory block is formed by multiple Z bunch, described N-memory block is formed by multiple N bunch, described Z-bunch and N-bunch is respectively provided with multiple pieces, each piece is made up of multiple sectors, it is characterised in that described synchronization write-back method includes:
Arrange one and write spatial cache, relatively and obtain grade system and need the data block A and the described difference result of calculation diff writing the data block A ' in buffer memory with identical start sector number of write, and the data block A ' write in buffer memory is replaced with data block A;
Adopting the result diff that each difference calculates by absolute fair algorithm to give weighted value, result diff difference calculated according to weighted value is from significantly little sequence;
When there being general data block B request write, control data block B is compressed, save segment space, the result diff of the difference calculating that weighting weight is maximum is inserted into described segment space and forms data block B ', and to control to be written to data block B ' first block of disk be in empty Z-bunch, and data block B is cached to writes in spatial cache;
When the described utilization rate writing buffer memory reaches predetermined threshold value, displace the data block on the data block Replace Disk and Press Anykey To Reboot write in buffer memory, make Z-bunch to become N bunch.
2. synchronization write-back method according to claim 1, it is characterized in that, described arrange one and write buffer memory, relatively and obtain grade system and need the concretely comprising the following steps of data block A and the described difference result of calculation diff writing the data block A ' in buffer memory with identical start sector number of write:
Shelves system receives write request, the start sector number of the described data block A in acquisition write request, and searches in writing buffer memory, if there is the data block A ' of identical start sector number;
If it is present use difference to calculate, calculate the difference result of calculation diff between data block A and data block A ';
If it does not exist, then directly write request is submitted in request queue.
3. synchronization write-back method according to claim 2, it is characterised in that when there being new general data block B request write, further comprised the steps of: before controlling data block B is compressed
Judge whether variant result of calculation diff;
If variant result of calculation diff, partial block pattern persistent storage data mode is then adopted first data block B to be compressed, save segment space, calculate the size Len of the size data block B after compression, and the result diff that the heavy maximum difference of weighting calculates is inserted in described saved segment space and forms data block B ', and data block B ' is written to described disk;
If zero difference result of calculation diff, adopt full block pattern persistent storage data mode to update disk logical structure, then described data block B is written in the block after first data block of described Z race.
4. synchronization write-back method according to claim 3, it is characterised in that the result diff of the difference calculating that described weighting weight is maximum is inserted into concretely comprising the following steps in described saved segment space:
Difference result of calculation diff is compressed, and uses timestamp to identify this difference result of calculation diff, then the difference result of calculation diff with timestamp is embedded in described saved segment space and forms data block B ';
Again write request is submitted in request queue, and data block B is cached to writes in buffer memory, and the read operation service of these part data is provided.
5. synchronization write-back method according to claim 3, it is characterized in that, described partial block pattern persistent storage data mode is in disk, the data block write in buffer memory providing the read operation of these part data by data block B ' storage with variant result of calculation diff in request queue.
6. synchronization write-back method according to claim 1, it is characterised in that described displace the data block on the data block Replace Disk and Press Anykey To Reboot write in buffer memory when the described utilization rate writing buffer memory reaches predetermined threshold value, makes Z-bunch to become N bunch concretely comprise the following steps:
If writing utilization rate in buffer memory to have reached threshold value, then use replacement algorithm, select the some data block replacement write in spatial cache to go out, by full block pattern persistent storage data mode, this data block is covered the data block B ' of previous employing partial block pattern perdurable data storage mode storage.
7. synchronization write-back method according to claim 1, it is characterised in that described synchronization write-back method also includes:
For the read request that system is submitted to, in writing buffer memory, search whether to meet the data of requirement;
If it has, then directly take out these data to provide reading service;
Otherwise, read request is ranked up, is then distributed in read request queue from high to low by weight, then read data to disk, it is provided that the reading service of shelves system.
8. synchronization write-back method according to claim 1, it is characterized in that, described synchronization write-back method also includes: after system reboot, by scanning disk, search embedded in the data block of difference result of calculation diff, and these data blocks are loaded into write in spatial cache, the read operation of these part data is provided for shelves system.
9. synchronization write-back method according to claim 1, it is characterised in that the size respectively 64MB of described Z-memory block and N-memory block, the size of described Z bunch and N bunch respectively 64KB, described piece be sized to 4KB.
10. synchronization write-back method according to claim 1, it is characterised in that described arrange and write concretely comprising the following steps of buffer memory:
One piece of write-back memory space is distributed, for buffer unit divided data block, it is provided that the read operation of these data for I/O scheduler.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510976637.9A CN105630413B (en) | 2015-12-23 | 2015-12-23 | A kind of synchronization write-back method of data in magnetic disk |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510976637.9A CN105630413B (en) | 2015-12-23 | 2015-12-23 | A kind of synchronization write-back method of data in magnetic disk |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105630413A true CN105630413A (en) | 2016-06-01 |
| CN105630413B CN105630413B (en) | 2018-09-28 |
Family
ID=56045411
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510976637.9A Active CN105630413B (en) | 2015-12-23 | 2015-12-23 | A kind of synchronization write-back method of data in magnetic disk |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105630413B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106604111A (en) * | 2016-12-16 | 2017-04-26 | 深圳市九洲电器有限公司 | Set-top box Flash data storage method and set-top box Flash data storage system |
| CN109144418A (en) * | 2018-08-16 | 2019-01-04 | 郑州云海信息技术有限公司 | A kind of method of data synchronization and device |
| CN110780808A (en) * | 2018-07-31 | 2020-02-11 | 佳能株式会社 | Information processing apparatus, control method therefor, and storage medium |
| CN114064588A (en) * | 2021-11-24 | 2022-02-18 | 建信金融科技有限责任公司 | Storage space scheduling method and system |
| CN117850705A (en) * | 2024-02-22 | 2024-04-09 | 北京壁仞科技开发有限公司 | Artificial intelligent chip and data synchronization method thereof |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1704910A (en) * | 2004-06-03 | 2005-12-07 | 华为技术有限公司 | Write handling method for disc array arrangement |
| US7908448B1 (en) * | 2007-01-30 | 2011-03-15 | American Megatrends, Inc. | Maintaining data consistency in mirrored cluster storage systems with write-back cache |
| CN102567445A (en) * | 2011-10-25 | 2012-07-11 | 无锡城市云计算中心有限公司 | Method for guaranteeing consistency of metadata in distributed file system |
| CN103064929A (en) * | 2012-12-24 | 2013-04-24 | 创新科存储技术（深圳）有限公司 | Method for server writing data in network file system |
-
2015
- 2015-12-23 CN CN201510976637.9A patent/CN105630413B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1704910A (en) * | 2004-06-03 | 2005-12-07 | 华为技术有限公司 | Write handling method for disc array arrangement |
| US7908448B1 (en) * | 2007-01-30 | 2011-03-15 | American Megatrends, Inc. | Maintaining data consistency in mirrored cluster storage systems with write-back cache |
| CN102567445A (en) * | 2011-10-25 | 2012-07-11 | 无锡城市云计算中心有限公司 | Method for guaranteeing consistency of metadata in distributed file system |
| CN103064929A (en) * | 2012-12-24 | 2013-04-24 | 创新科存储技术（深圳）有限公司 | Method for server writing data in network file system |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106604111A (en) * | 2016-12-16 | 2017-04-26 | 深圳市九洲电器有限公司 | Set-top box Flash data storage method and set-top box Flash data storage system |
| CN110780808A (en) * | 2018-07-31 | 2020-02-11 | 佳能株式会社 | Information processing apparatus, control method therefor, and storage medium |
| CN110780808B (en) * | 2018-07-31 | 2023-10-31 | 佳能株式会社 | Information processing apparatus and control method thereof |
| CN109144418A (en) * | 2018-08-16 | 2019-01-04 | 郑州云海信息技术有限公司 | A kind of method of data synchronization and device |
| CN114064588A (en) * | 2021-11-24 | 2022-02-18 | 建信金融科技有限责任公司 | Storage space scheduling method and system |
| CN114064588B (en) * | 2021-11-24 | 2023-04-25 | 建信金融科技有限责任公司 | Storage space scheduling method and system |
| CN117850705A (en) * | 2024-02-22 | 2024-04-09 | 北京壁仞科技开发有限公司 | Artificial intelligent chip and data synchronization method thereof |
| CN117850705B (en) * | 2024-02-22 | 2024-05-07 | 北京壁仞科技开发有限公司 | Artificial intelligent chip and data synchronization method thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105630413B (en) | 2018-09-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106716334B (en) | Efficient data movement within a file system volume | |
| CN105630413A (en) | Synchronized writeback method for disk data | |
| US8909887B1 (en) | Selective defragmentation based on IO hot spots | |
| US7058783B2 (en) | Method and mechanism for on-line data compression and in-place updates | |
| US6988165B2 (en) | System and method for intelligent write management of disk pages in cache checkpoint operations | |
| US8402205B2 (en) | Multi-tiered metadata scheme for a data storage array | |
| US9779027B2 (en) | Apparatus, system and method for managing a level-two cache of a storage appliance | |
| US6968424B1 (en) | Method and system for transparent compressed memory paging in a computer system | |
| US8127084B2 (en) | Using different algorithms to destage different types of data from cache | |
| US9355112B1 (en) | Optimizing compression based on data activity | |
| US7721043B2 (en) | Managing write requests in cache directed to different storage groups | |
| CN108604165B (en) | Storage device | |
| CN102023809A (en) | Storage system, method for reading data from storage system and method for writing data to storage system | |
| US9727245B2 (en) | Method and apparatus for de-duplication for solid state disks (SSDs) | |
| CN107924291B (en) | Storage system | |
| CN103246696A (en) | High-concurrency database access method and method applied to multi-server system | |
| KR20100079654A (en) | Methods of distributing log block associativity in real-time system and flash memory device performing the same | |
| CN102087586A (en) | Data processing method and device | |
| CN101644996A (en) | Storage method of index data and storage control device | |
| CN107423229B (en) | Buffer area improvement method for page-level FTL | |
| CN115427941A (en) | Data management system and control method | |
| EP3255547B1 (en) | Data backup device and method | |
| US10489296B2 (en) | Quality of cache management in a computer | |
| CN109739696B (en) | Double-control storage array solid state disk caching acceleration method | |
| CN104679684B (en) | Semiconductor devices and its operating method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |