CN103714010B - Storage device write-in method and storage device - Google Patents
Storage device write-in method and storage device Download PDFInfo
- Publication number
- CN103714010B CN103714010B CN201310724721.2A CN201310724721A CN103714010B CN 103714010 B CN103714010 B CN 103714010B CN 201310724721 A CN201310724721 A CN 201310724721A CN 103714010 B CN103714010 B CN 103714010B
- Authority
- CN
- China
- Prior art keywords
- memory block
- raid
- data
- write
- address
- 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.)
- Active
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention discloses a storage device write-in method and a storage device. The storage device write-in method includes the steps of firstly, obtaining attributes of data to be written in; secondly, if the data are confirmed as first data according to the attributes of the data to be written in, writing the first data into storage blocks which belong to a first RAID according to the characteristic that an SSD rapidly makes a response, wherein the total storage space size value of the first RAID is equal to the volume of at least one SSD; thirdly, if the data are confirmed as second data according to the attributes of the data to be written in, writing the second data into storage blocks which belong to a second RAID, wherein the total storage space size value of the second RAID is equal to the sum of the volumes of data disks. By means of the storage device write-in method and the storage device, the write-in speed of the storage device can be increased after sectors are improved, and therefore the performance of the storage device is improved, and the service life of the storage device is prolonged.
Description
Technical field
The present embodiments relate to memory technology, more particularly to a kind of storage device wiring method and storage device.
Background technology
With the continuous development of computer technology and the communication technology, the application of hard disk and demand are more and more.Computer is only
The data stored in conventional hard can be read and write in units of sector, the sector-size of conventional hard is 512 words
Section, i.e. 512B.In recent years, sector is increased to 4096 bigger, more efficient words by hard disk manufacturing business from 512 traditional bytes
Section, i.e. 4KB, will 8 byte sectors of tradition 512 merge into a 4KB sector, the i.e. hard disk and can be carried out in units of 4KB
Read and write, it is possible to increase areal concentration and hard-disk capacity, while providing more powerful error correcting function.
But, many aspects of computer are still given tacit consent to hard disk is read and writen in units of 512B.Although meter
Calculate it is machine-readable take less than 4KB data when do not result in what affect, but when computer attempt write new data be less than 4KB
When, when the data of such as 2 512B write hard disk, need to read the fans of the whole 4KB comprising computer write request target location first
The data in area, then replace new data by the data available data of equal size, such as replace the legacy data of 2 512B, will be new
Data and the available data of 6 512B do not replaced of 2 512B merge into the data of a 4KB, could again by this 4KB
Data write hard disk, and, because existing hard disk mostly is mechanical hard disk(HardDiskDrive, data disks)Storage, data disks
Response speed it is slower, cause the speed for writing hard disk very slow, cause hard disk performance poor, and polylith hard disk is put when being used together
For controlling hard disk disk, it is ensured that data can correctly read and write array and in turn result in the decline of hard disk overall performance.Overall performance under
Drop, meanwhile, frequently reading writing harddisk, the magnetic head that will also result in hard disk ceaselessly changes, also can reduce the life-span of hard disk, therefore carries
The service life of high hard disk writing speed and lifting hard disk becomes problem demanding prompt solution.
The content of the invention
The embodiment of the present invention provides a kind of storage device wiring method and storage device, can accelerate depositing after sector is lifted
Storage equipment writing speed, so as to lift the performance of storage device, improves the service life of storage device.
The embodiment of the present invention provides a kind of storage device wiring method, including:
Obtain the attribute of data to be written;
If determining that the data are the first data according to the attribute of the data to be written, using solid-state hard disk SSD
The characteristic of quick response belongs to first data write in the memory block of the first RAID, and total storage of a RAID is empty
Between capacity of the sizes values equal at least about SSD;If determining that the data are second according to the attribute of the data to be written
Data, then belong to second data write in the memory block of the 2nd RAID, total storage size of the 2nd RAID
Capacity sum of the value equal to total data disks.
In the first possible implementation, according in a first aspect, before the attribute for obtaining data to be written,
Also include:
Determine according to the storage size value of a RAID and the sizes values of the memory block and belong to described first
The memory block number of RAID is M, and the sizes values of the storage size value according to the 2nd RAID and the memory block
It is determined that belong to the memory block number of the 2nd RAID for N number of, so that being labeled as belonging to the memory block number of a RAID not
More than M, the memory block number for being labeled as belonging to the 2nd RAID is not more than N;
Wherein, the storage size value of a RAID determines according to the capacity of at least one SSD, described the
The storage size value of two RAID determines that the sizes values of the memory block are default for one according to the capacity of total data disks
Definite value, described M, N be positive integer.
It is described by institute with reference to first aspect and the first possible implementation in second possible implementation
Stating the write of the first data and belonging to the memory block of a RAID includes:
Determine that first data need to write the memory block for belonging to a RAID;
According to all of memory block of default table item index, the address of all memory blocks for being labeled as belonging to a RAID is obtained
And the address of all unlabelled memory blocks;
First data are write in corresponding memory block according to the address of the memory block.
In the third possible implementation, with reference to first aspect, the first possible implementation and second can
The mode that can be realized, it is described to include the memory block that second data write belongs to the 2nd RAID:
Determine that second data need to write the memory block for belonging to the 2nd RAID;
According to all of memory block of default table item index, the address of all memory blocks for being labeled as belonging to the 2nd RAID is obtained
And the address of all unlabelled memory blocks;
Second data are write in corresponding memory block according to the address of the memory block.
In the 4th kind of possible implementation, with reference to first aspect, the first possible implementation, second possibility
The mode of realization and the third mode in the cards, before the sizes values that acquisition data to be written take up room, also
Including:
If the memory block for belonging to a RAID is otherwise fully engaged, by it is described be otherwise fully engaged belong to depositing for a RAID
Data conversion storage in storage block is in the memory block of the 2nd RAID.
Second aspect, a kind of storage device provided in an embodiment of the present invention, including:
Acquiring unit, for obtaining the attribute of data to be written;
Write control unit, if for determining that the data are the first data according to the attribute of the data to be written,
Then first data write is belonged in the memory block of a RAID using the characteristic of solid-state hard disk SSD quick response, it is described
Capacity of the total storage size value of the first RAID equal at least about a SSD;If according to the attribute of the data to be written
Determine that the data are the second data, then belong to second data write in the memory block of the 2nd RAID, described second
Total storage size value of RAID is equal to the capacity sum of total data disks.
In the first possible implementation, according in a first aspect,
Processing unit, for determining according to the sizes values of the storage size value of a RAID and the memory block
The memory block number for belonging to a RAID is M, and the storage size value according to the 2nd RAID and described is deposited
The sizes values of storage block determine that the memory block number for belonging to the 2nd RAID is N number of, so that being labeled as belonging to a RAID's
Memory block number is not more than M, and the memory block number for being labeled as belonging to the 2nd RAID is not more than N;
Wherein, the storage size value of a RAID determines according to the capacity of at least one SSD, described the
The storage size value of two RAID determines that the sizes values of the memory block are default for one according to the capacity of total data disks
Definite value, described M, N be positive integer.
In second possible implementation, with reference to first aspect and the first possible implementation, said write
Control unit includes:
Determining module, for determining that first data need to write the memory block for belonging to a RAID;
Acquisition module, for according to all of memory block of default table item index, acquisition is all to be labeled as belonging to a RAID
Memory block address and the address of all unlabelled memory blocks;
Writing module, for first data to be write in corresponding memory block according to the address of the memory block.
In the third possible implementation, with reference to first aspect, the first possible implementation and second can
The mode that can be realized, the determining module is additionally operable to determine that second data need to write the memory block for belonging to the 2nd RAID;
The determining module, is additionally operable to determine that second data need to write the memory block for belonging to the 2nd RAID;
The acquisition module, is additionally operable to according to the default all of memory block of table item index, obtains and all is labeled as belonging to the
The address of the memory block of two RAID and the address of all unlabelled memory blocks;
Said write module, is additionally operable to for second data to write corresponding memory block according to the address of the memory block
In.
In the 4th kind of possible implementation, with reference to first aspect, the first possible implementation, second possibility
The mode of realization and the third mode in the cards,
Said write control unit, if the memory block for being additionally operable to belong to a RAID is otherwise fully engaged, will be described complete
Data conversion storage in the full memory block for belonging to a RAID for taking is in the memory block of the 2nd RAID.
Storage device wiring method provided in an embodiment of the present invention and storage device.Storage device obtains data to be written
Attribute;If determining that data are the first data according to the attribute of data to be written, using solid-state hard disk SSD quick response
Characteristic is write first data in the memory block for belonging to a RAID, and total storage size value of a RAID is equal at least about one
The capacity of individual SSD;If determining data for the second data according to the attribute of data to be written, the write of the second data is belonged to into the
In the memory block of two RAID, total storage size value of the 2nd RAID is equal to the capacity sum of total data disks.So,
Storage device can utilize the data that the fast response characteristic of SSD is little input and output IO to attribute to write, and be write with improving
The speed for entering, and data disks only process the data of big IO, data disks can will be big equal to sector in the data of this kind of big IO
The data of little value are write direct, and with this speed of write is accelerated, and are reduced because the magnetic head of hard disk frequently reads and writes the change for bringing, and are lifted
The service life of hard disk.Storage device is this while can accelerate fan using the structure and storage mode of SSD and data disk storage
Writing speed of storage equipment after area's lifting, and then lift the performance of storage device.
Description of the drawings
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
The accompanying drawing to be used needed for having technology description is briefly described, it should be apparent that, drawings in the following description are these
Some bright embodiments, for those of ordinary skill in the art, without having to pay creative labor, can be with
Other accompanying drawings are obtained according to these accompanying drawings.
The flow chart of the storage device wiring method embodiment that Fig. 1 is provided for the present invention;
The flow chart of storage device wiring method another embodiment that Fig. 2 is provided for the present invention
Fig. 3 is the structural representation of hard disk theory of constitution;
The flow chart of the another embodiment of storage device wiring method that Fig. 4 is provided for the present invention;
The structural representation of the storage device embodiment that Fig. 5 is provided for the present invention;
The structural representation of another embodiment of storage device that Fig. 6 is provided for the present invention;
The structural representation of the another embodiment of storage device that Fig. 7 is provided for the present invention;
The structural representation of another storage device embodiment that Fig. 8 is provided for the present invention.
Specific embodiment
To make purpose, technical scheme and the advantage of the embodiment of the present invention clearer, below in conjunction with the embodiment of the present invention
In accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is
The a part of embodiment of the present invention, rather than the embodiment of whole.Based on the embodiment in the present invention, those of ordinary skill in the art
The every other embodiment obtained under the premise of creative work is not made, belongs to the scope of protection of the invention.
The flow chart of the storage device wiring method embodiment that Fig. 1 is provided for the present invention, the storage device is by least one
Solid state hard disc(Solid State Disk, SSD)Combine to be formed with traditional data disks, the write-in functions of hard disk are by SSD sums
Support to realize jointly according to disk, as shown in figure 1, the method for the present embodiment can include:
S101, storage device obtain the attribute of data to be written.
For example, storage device can be hard disk or other equipment with store function, and the present embodiment is with hard disk act
Example is illustrated, because there is the major part of hard disk sector computer the functional area of write demand still to give tacit consent to the size of sector
It is worth for traditional 512B, therefore, still write in units of 512B when data are write to hard disk, that is to say, that to be written
The sizes values for taking up room of the data for entering can be the integral multiple of 512B, such as 3 512B, 2 512B, 10 512B, 19
512B etc..
It should be noted that prior art can lift the sizes values of the sector of hard disk to being 8 with 4KB
512B, that is to say, that hard disk can be read in units of 4KB and is written with, so the embodiment of the present invention is with the size of sector
It is worth and is illustrated for 4KB citings, write of any sector-size value more than the sizes values that the data of computer write take up room
Within the scope of the present invention, the such as sizes values of sector are 8KB, computer still with 512B writes, or, sector it is big
Little value is 8KB, and upgrading computer with 4KB to carrying out writing etc. within protection domain.
Further, the attribute of data can include big input and output (Input Output, IO) and little IO, wherein, IO
Size can be divided based on the size of a slitting, less than a slitting size IO be little IO, more than or equal to one point
The IO of bar size is big IO;One slitting size=(number -1 of the disk of RAID groups) * segment depth, slitting is same
In RAID groups, physical deflection identical several stripe units composition, segment depth be the block in a stripe unit
Number, block is the least unit of storage device read-write operation, and size is 512 bytes, and stripe unit is continuous on same hard disk
The element of the slitting of several block compositions.
If S102, storage device determine data for the first data, execution step S103 according to the attribute of data to be written;
If determining data for the second data, execution step S104 according to the attribute of data to be written.
For example, if storage device determines the data for little IO, execution step according to the attribute of data to be written
S103, however, it is determined that the data are big IO then execution steps S104.
Further, confirm that IO can if the RAID component bar sizes are 1M, but be write for big IO or little IO
IO be 10K, in fact at this time the attribute of IO should be determined as little IO, at this time just this 10K IO be write completely first
RAID;If slitting size is 100K, the I O size to be write is 101K, then this when just all writes the data of this 101K
In 2nd raid.
S103, storage device write first data into the memory block for belonging to a RAID using the characteristic of SSD quick responses
In, total storage size value of a RAID is equal at least about the capacity of a SSD.
Further, during the data of the sizes values that computer write is less than sector, such as but sector is 4KB number to be written
When according to the sizes values for taking up room being 3 512B, first reading computer write request target location, to be that SSD is corresponding belong to first
The data of one whole 4KB sectors of the memory block of RAID, then replace the legacy data of this 4KB data 3 512B sizes by
For new data, i.e., legacy data is covered with new data, the data of 3 for being written into 512B show with 5 512B's not replaced
There are data to merge into the data of a new 4KB, then the data write of this 4KB is belonged to into the memory block of a RAID.Due to
One RAID is supported by least one SSD, and SSD has the characteristic of quick response, you can with quickly reading and write diagnostics, can
To improve the writing speed of little IO.Time such as the data write conventional hard of 3 512B may be 1 second, but 3 512B
The time of the memory block of the RAID that data write SSD is supported may only be 10 milliseconds, greatly improve small data input IO
The time of write.
S104, storage device belong to the write of the second data in the memory block of the 2nd RAID, and total storage of the 2nd RAID is empty
Between sizes values be equal to total data disks capacity sum.
For example, when the data attribute of computer write is big IO, the data division of the sizes values of sector is equal in IO
Just write direct in the 2nd RAID, such as sector is 4KB, when the sizes values that big I/O data to be written takes up room are 8 512B,
Directly the data to be written write that this big I/O data is divided into two 4KB can be belonged in the memory block of the 2nd RAID, and not
Needs first read again from the 2nd RAID replaces again the data for finally writing this 2 4KB again, saves the time of write data.
If the sizes values that big I/O data to be written takes up room are more than 4K, and are unsatisfactory for the situation of the integral multiple of IO
Under, such as during 15K, 15K is done a rounding by 4K, such as can be processed with the data that rounding is a 15K, i.e. 15K will write
How many 4K units, only pre-read the part for being unsatisfactory for 4K, and the such as data of 15K roundings necessarily have a part less than 4K.If
Will be from address block(block)Where 0, a block as described above is 512B, and the data for writing a 15K are gone down, at this time
Its real address 0 is alignment, and data 15K to be write are not lined up, but is also that last 4k is not lined up, and at this time, is just only needed
Pre-read in last 4K data, do and merge, then write;And if to be to go to write a 15K where 1 from address block
Data, at this time address and size of data 15K are not 4K alignment, it is necessary to address and data size all registration process,
And in fact simply the parts of 4K end to end of this 15K data do not line up yet, 0 to 4K and 12K to 16K this part number is at this time just pre-read
According to, do and merge, then write.
It should be noted that although the present embodiment does not do any restriction with hard disk illustration with this, other storages set
From being caused according to the unit write after the lifting when standby write unit is lifted and computer or user equipment is write
Write it is slow using method provided in an embodiment of the present invention raising writing speed, so within protection domain.
Storage device wiring method provided in an embodiment of the present invention.Storage device obtains the attribute of data to be written;If
Data are determined for the first data according to the attribute of data to be written, then using the characteristic of solid-state hard disk SSD quick response by the
The write of one data belongs in the memory block of a RAID, and total storage size value of a RAID is equal at least about SSD
Capacity;If determining that data are the second data according to the attribute of data to be written, the write of the second data is belonged to into the 2nd RAID
Memory block in, total storage size value of the 2nd RAID is equal to the capacity sum of total data disks.So, storage sets
It is standby to be write using the data that the fast response characteristic of SSD is little input and output IO to attribute, to improve the speed of write
Degree, and data disks only process the data of big IO, data disks can be by the sizes values equal to sector in the data of this kind of big IO
Data are write direct, and with this speed of write is accelerated, and are reduced because the magnetic head of hard disk frequently reads and writes the change for bringing, and lift hard disk
Service life.Storage device is this to be lifted while can accelerate sector using the structure and storage mode of SSD and data disk storage
Writing speed of storage equipment afterwards, and then lift the performance of storage device.
The flow chart of storage device wiring method another embodiment that Fig. 2 is provided for the present invention, the present embodiment is storing
Equipment is that hard disk is illustrated, but with this does not do any restriction.As shown in Fig. 2 the method includes:
S201, hard disk belong to a RAID according to the storage size value of a RAID and the sizes values determination of memory block
Memory block number be M, and the sizes values of the storage size value according to the 2nd RAID and memory block determine and belong to second
The memory block number of RAID is N number of.
Further, the memory block number for being labeled as belonging to a RAID is caused to be not more than M by calculating memory block number,
The memory block number for being labeled as belonging to the 2nd RAID is not more than N.
Wherein, the storage size value of a RAID is according to the determination of the capacity of at least one SSD, the storage of the 2nd RAID
Space size value determines that the sizes values of memory block are a default definite value, and M, N are positive integer according to the capacity of total data disks.
For example, the storage size value of a RAID determines it can is basis according to the capacity of at least one SSD
RAID is determining the space size of RAID groups.If SSD is composition RAID10 ranks, so, the capacity of raid groups here
Should be 32GB, in 2 pieces of disks, have 1 piece of disk to be for doing data protection, that is to say, that the data on 2 SSD disks.
And it can be according to data disks that the sizes values of the storage size value according to the 2nd RAID and memory block determine
Number and the RAID being located determine that the termination capacity for being primarily due to storage device is the number with data disks and RAID groups
Rank it is related.Total capacity such as the 2nd RAID is equal to the capacity sum of total data disks:If RAID group ranks are
RAID10, then be multiplied by mirror set number equal to the capacity of data disks;If RAID group ranks are RAID3 or RAID5, equal to into
Member's disk number deducts 1, and the capacity of each disk is then multiplied by again;If RAID group ranks are RAID6, equal to member data disk
Number deducts 2, is multiplied by the capacity of each disk.Due to the capacity of each data disks it is different, therefore the capacity for being multiplied by data disks when
Wait, be defined by the capacity of that minimum disk of capacity in all disks constituted in RAID groups and calculated.
Such as, according to the above-mentioned storage size value for being calculated a RAID be 32GB, the memory space of the 2nd RAID
Sizes values are 128GB, and the sizes values of memory block are a default definite value, and such as 64M then can determine and belong to depositing for a RAID
Storage block number M=512, belongs to memory block number N=2048 of the 2nd RAID.
That is, when being subsequently marked to unlabelled memory block will with belong to a RAID memory block number not
More than 512, be not more than 2048 with the memory block number for belonging to the 2nd RAID is carried out as standard.Such as, the memory block of a RAID
1 is labeled as using a flag bit, the memory block of the 2nd RAID is labeled as 0 using a flag bit, if being then labeled as 1 storage
Block number has equalized to 512, then unlabelled other memory blocks are no longer labeled as 1, in the same manner, if being labeled as 0 memory block number
2048 are had equalized to, then unlabelled other memory blocks are no longer labeled as 0.
It should be noted that the capacity of above-mentioned SSD and data disks is only used for illustrating, the sizes values and mark of memory block
Mode is also illustration, does not do any restriction with this, and other numerical value and mark mode are within protection domain.
S202, hard disk will then belong to depositing for a RAID if it is determined that the memory block that M belongs to a RAID is otherwise fully engaged
Data in storage block are all dumped in the memory block of the 2nd RAID, and are emptied in M memory block for belonging to a RAID of occupancy
All data.
It should be noted that if 512 memory blocks for belonging to a RAID are otherwise fully engaged, i.e., each memory block is filled
Data are expired, and also there is vacant memory block and being labeled as in hard disk and belong to the 2nd RAID's but the storage that is not otherwise fully engaged
Block, then hard disk can at one's leisure between, i.e., when not reading and write any data, by the memory block of a RAID
Data are all dumped in the memory block or vacant memory block of the 2nd RAID, and proceed to needs to deposit vacant after vacant memory block
Storage block is labeled as belonging to the 2nd RAID.
S203, hard disk obtain the attribute of data to be written.
It should be noted that it can be that the data are little IO or big I/O data that acquisition is attribute, in above-described embodiment
Jing is described, and be will not be described here.
If S204, hard disk determine data for the first data, execution step S205 according to the attribute of data to be written;If root
According to the attribute of data to be written data are determined for the second data, execution step S208.
S205, hard disk determine that the first data need to write the memory block for belonging to a RAID.
Further, hard disk can utilize the characteristic of SSD quick responses to complete step S205~S207.
, according to all of memory block of default table item index, acquisition is all to be labeled as belonging to depositing for a RAID for S206, hard disk
The address of storage block and the address of all unlabelled memory blocks.
S207, hard disk write the first data in corresponding memory block according to the address of memory block.
For example, Fig. 3 for hard disk theory of constitution structural representation, as shown in figure 3, total memory space of hard disk can be with
It is that SSD and data disks are constituted together, SSD and data disks realize store function in bottom, and no longer as prior art
Caching is provided by SSD, by data disk storage.Such as a RAID can be constituted by 2 SSD, be then made up of 3 data disks
2nd RAID.Hard disk to a RAID and the 2nd RAID unified addressings, can distinguish the attribute of the data that user need to write, that is, count
According to sizes values, when data sizes values be less than sector, when being little I/O data, correspondence the first RAID address just can be by little IO
Data write SSD, and when for more than or equal to the data of the sizes values of sector, the address of the 2nd RAID of correspondence just can will be this kind of
Larger data write data disks.
Table 1 is default list item, there is two-layer MBA memory block address in hard disk(Lba, Logical Block Address), one
The individual LUN for corresponding to real memory space(Logical Unit Numbe, LUN), each true Lba is
The Lba of LUN is 64M, and another is virtual Lba, is denoted as the Lba of block BLOCK.As shown in table 1, hard disk is by SSD and data disks
Total memory space of composition is mapped to the Lba of the Lba and BLOCK of LUN, and the Lba to LUN is numbered, the Lba of the LUN with
The Lba of BLOCK can mutually be uniquely determined with address.The Lba of each LUN can also be marked in this list item to be belonging to first
RAID or the 2nd RAID, can such as utilize the flag bit in the example above to be labeled as belonging to a RAID or second for 0 or 1
RAID。
When data to be written are the first data, look for whether having depositing for the 64M sizes for belonging to a RAID first
Storage block is used for storing first data, if it has, judging whether the memory block is filled with again;If be filled with, go to obtain
What other were not filled with belongs to the memory block of a RAID, if all memory blocks for belonging to a RAID are filled with, finds idle
64M sizes memory block, and the memory block is labeled as to belong to a RAID.Further, the RAID of SSD compositions can
With difference, e.g., polylith SSD composition RAID3 or RAID6 etc..
Table 1
Further, there can be preset table in hard disk, the form can increase to have each memory block on the basis of table 1
Address.Hard disk can be according to all of memory block of default table item index, using unlabelled memory block as first memory block, such as
The memory block of Serial No. 3 is first memory block, will be otherwise fully engaged in all memory blocks for being labeled as and belong to a RAID
Memory block be defined as the second memory block, the such as memory block of Serial No. 1 is otherwise fully engaged, and the memory block of Serial No. 5 is not
It is otherwise fully engaged, then the memory block of Serial No. 5 is labeled as into the second memory block, the address of the second memory block is obtained, by first
Data are write in the second memory block according to the address of the second memory block.Memory block such as the memory block of Serial No. 1 is accounted for completely
With the memory block of Serial No. 5 is also otherwise fully engaged, then obtain the address of first memory block, and the first data are deposited according to first
In the address write first memory block of storage block, and first memory block is marked to belong to a RAID, i.e., in the memory block of Serial No. 3
A RAID hurdles in mark 1.
S208, hard disk determine that the second data need to write the memory block for belonging to the 2nd RAID.
, according to all of memory block of default table item index, acquisition is all to be labeled as belonging to depositing for the 2nd RAID for S209, hard disk
The address of storage block and the address of all unlabelled memory blocks.
S210, hard disk write the second data in corresponding memory block according to the address of memory block.
For example, the preset table in hard disk can increase the address for having each memory block on the basis of table 1.Hard disk
Can be according to all of memory block of default table item index, using unlabelled memory block as first memory block, such as Serial No. 3
Memory block is first memory block, and the memory block not being otherwise fully engaged in all memory blocks for being labeled as and belong to the 2nd RAID is determined
For the 3rd memory block, the such as memory block of Serial No. 2 is otherwise fully engaged, and the memory block of Serial No. 4 is not otherwise fully engaged, then
The memory block of Serial No. 4 is labeled as into the 3rd memory block, the address of the 3rd memory block is obtained, the second data are deposited according to the 3rd
The address of storage block is write in the 3rd memory block.Memory block such as the memory block of Serial No. 2 is otherwise fully engaged, Serial No. 4
Memory block is also otherwise fully engaged, then obtain the address of first memory block, and the second data are write according to the address of first memory block
In first memory block, and first memory block is marked to belong to the 2nd RAID, i.e., in the 2nd RAID hurdles of the memory block of Serial No. 3
Mark 0.
Storage device wiring method provided in an embodiment of the present invention.Storage device obtains the attribute of data to be written;If
Data are determined for the first data according to the attribute of data to be written, then using the characteristic of solid-state hard disk SSD quick response by the
The write of one data belongs in the memory block of a RAID, and total storage size value of a RAID is equal at least about SSD
Capacity;If determining that data are the second data according to the attribute of data to be written, the write of the second data is belonged to into the 2nd RAID
Memory block in, total storage size value of the 2nd RAID is equal to the capacity sum of total data disks.So, storage sets
It is standby to be write using the data that the fast response characteristic of SSD is little input and output IO to attribute, to improve the speed of write
Degree, and data disks only process the data of big IO, data disks can be by the sizes values equal to sector in the data of this kind of big IO
Data are write direct, and with this speed of write is accelerated, and are reduced because the magnetic head of hard disk frequently reads and writes the change for bringing, and lift hard disk
Service life.Storage device is this to be lifted while can accelerate sector using the structure and storage mode of SSD and data disk storage
Writing speed of storage equipment afterwards, and then lift the performance of storage device.
The flow chart of the another embodiment of storage device wiring method that Fig. 4 is provided for the present invention, the present embodiment is also with storage
Equipment is illustrated for hard disk, as shown in figure 4, it is with step that the method for the present embodiment is different from above-mentioned embodiment illustrated in fig. 3
Rapid S211 substituted for S202.
S211, hard disk then belong to if it is determined that the memory block for belonging to a RAID is otherwise fully engaged by what is be otherwise fully engaged
Data conversion storage in the memory block of one RAID is in the memory block of the 2nd RAID.
For example, if the memory block for having at least one the oneth RAID in hard disk is otherwise fully engaged, hard disk can be in sky
By the memory block of the data conversion storage in the memory block for belonging to a RAID being otherwise fully engaged to the 2nd RAID between idle.Hard disk can
Emptied to store other data with the RAID after data are produced.
Storage device wiring method provided in an embodiment of the present invention.Storage device obtains what data to be written took up room
Sizes values;If sizes values of the sizes values that data to be written take up room less than sector, it is determined that data are the first data, sharp
During the memory block for belonging to a RAID is write first data into the characteristic of SSD quick responses, a RAID is by least one SSD
Support;If the sizes values that data to be written take up room are not less than the sizes values of sector, it is determined that data are the second data, will
The write of second data belongs in the memory block of the 2nd RAID, and the 2nd RAID is supported by data disks.So, storage device can
The data of sizes values less than sector are write using the fast response characteristic of SSD, to improve the speed of write, and HHD
The data of the sizes values more than or equal to sector are only processed, data disks can be by the size equal to sector in this kind of larger data
The data of value are write direct, and with this speed of write is accelerated.The writing speed of storage equipment after sector is lifted can be accelerated, and then
Improve the performance of storage device.
The structural representation of the storage device embodiment that Fig. 5 is provided for the present invention, it should be understood that depositing shown in Fig. 5
Storage equipment 30 is only an example, and storage device 30 can be hard disk, and its structure can have than the more mistake shown in figure
Or less part, two or more parts can be combined, or can have different parts to configure.Shown in figure
The various parts for going out can be in the hardware including one or more signal transactings and/or including special IC, software or hard
Realize in the combination of part and software.As shown in figure 5, the storage device 30 includes:Acquiring unit 301 and write control unit 302.
Acquiring unit 301, for being used to obtain the attribute of data to be written.
Write control unit 302, if for being used to determine that data are the first data according to the attribute of data to be written,
During the memory block for belonging to a RAID is write first data into using the characteristic of solid-state hard disk SSD quick response, a RAID's
Capacity of total storage size value equal at least about a SSD;If determining that data are second according to the attribute of data to be written
Data, then belong to the write of the second data in the memory block of the 2nd RAID, and total storage size value of the 2nd RAID is equal to always
Data disks capacity sum.
Further, if the memory block that write control unit 302 is additionally operable to belong to a RAID is otherwise fully engaged, will be by
Data conversion storage in the memory block for belonging to a RAID for taking completely is in the memory block of the 2nd RAID.
The structural representation of another embodiment of storage device that Fig. 6 is provided for the present invention, as shown in fig. 6, storage device 30
On the basis of structure shown in Fig. 5 further, can also include:Processing unit 303.
Processing unit 303, processing unit, for according to the sizes values of the storage size value of a RAID and memory block
It is determined that the memory block number for belonging to a RAID is M, and the storage size value according to the 2nd RAID and memory block is big
Little value determines that the memory block number for belonging to the 2nd RAID is N number of, so that being labeled as belonging to the memory block number of a RAID not
More than M, the memory block number for being labeled as belonging to the 2nd RAID is not more than N;
Wherein, the storage size value of a RAID is according to the determination of the capacity of at least one SSD, the storage of the 2nd RAID
Space size value determines that the sizes values of memory block are a default definite value, and M, N are positive integer according to the capacity of total data disks..
If further write control unit 301 is additionally operable to M memory block for belonging to a RAID of the determination of processing unit 303
It is otherwise fully engaged, then all dumps to the data in the memory block for belonging to a RAID in the memory block of the 2nd RAID, and clearly
Sky takes all data in the M memory block for belonging to a RAID.
The structural representation of the another embodiment of storage device that Fig. 7 is provided for the present invention, as shown in fig. 7, storage device 30
On the basis of structure shown in Fig. 5 or Fig. 6, write control unit 302 specifically can include:Determining module 3021, acquisition module
3022 and writing module 3023.
Determining module 3021, for determining that the first data need to write the memory block for belonging to a RAID.
Acquisition module 3022, for according to all of memory block of default table item index, acquisition is all to be labeled as belonging to first
The address of the memory block of RAID and the address of all unlabelled memory blocks.
Writing module 3023, for the first data to be write in corresponding memory block according to the address of memory block.
For example, determining module 3021 can be according to all of memory block of default table item index, by unlabelled storage
Block is defined as the memory block not being otherwise fully engaged in all memory blocks for being labeled as and belong to a RAID as first memory block
Second memory block;If there is the second memory block, acquisition module 3022 obtains the address of the second memory block, writing module 3023, tool
Body is used to the first data are write in the second memory block according to the address of the second memory block;If there is no the second memory block, obtain
Module 3022 obtains the address of first memory block, and writing module 3023 can write the first data according to the address of first memory block
In entering first memory block, and first memory block is marked to belong to a RAID.
Or,
Determining module 3021, is additionally operable to determine that the second data need to write the memory block for belonging to the 2nd RAID.
Acquisition module 3022, is additionally operable to according to the default all of memory block of table item index, obtains and all is labeled as belonging to the
The address of the memory block of two RAID and the address of all unlabelled memory blocks.
Writing module 3023, is additionally operable to the second data be write in corresponding memory block according to the address of memory block.
For example, determining module 3021 can be according to all of memory block of default table item index, by unlabelled storage
Block is defined as the memory block not being otherwise fully engaged in all memory blocks for being labeled as and belong to the 2nd RAID as first memory block
3rd memory block;If there is the 3rd memory block, acquisition module 3022 obtains the address of the 3rd memory block, and writing module 3023 is by the
Two data are write in the 3rd memory block according to the address of the 3rd memory block;If there is no the 3rd memory block, acquisition module 3022 is obtained
The address of first memory block is taken, the second data are write first memory block by writing module 3023 according to the address of first memory block
In, and mark first memory block to belong to the 2nd RAID.
The storage device of the present embodiment, can be used for performing the technical scheme of embodiment of the method shown in Fig. 1~Fig. 4, in fact
Existing principle is similar with technique effect, and here is omitted.
Storage device provided in an embodiment of the present invention.Storage device obtains the attribute of data to be written;If according to be written
The attribute of the data for entering determines that data are the first data, then write the first data using the characteristic of solid-state hard disk SSD quick response
In entering the memory block for belonging to a RAID, total storage size value of a RAID is equal at least about the capacity of a SSD;If root
Determine that data are the second data according to the attribute of data to be written, then the write of the second data is belonged to the memory block of the 2nd RAID
In, total storage size value of the 2nd RAID is equal to the capacity sum of total data disks.So, storage device being capable of profit
It is that the data of little input and output IO write to attribute with the fast response characteristic of SSD, to improve the speed of write, and data
Disk only processes the data of big IO, and data disks can directly write the data of the sizes values equal to sector in the data of this kind of big IO
Enter, the speed of write is accelerated with this, reduce because the magnetic head of hard disk frequently reads and writes the change for bringing, lift the service life of hard disk.
Storage device is this to be set while can be accelerated the storage after sector is lifted using the structure and storage mode of SSD and data disk storage
Standby writing speed, and then lift the performance of storage device.
The structural representation of another storage device embodiment that Fig. 8 is provided for the present invention, as shown in figure 8, with storage device
As a example by 40 are made up of data disks and at least one SSD, storage device 40 includes:Processor 401 and memory 402.
Processor 401, for obtaining the attribute of data to be written.
If memory 402 is used to determine that data are the first data according to the attribute of data to be written, using solid
The characteristic of state hard disk SSD quick responses is write first data in the memory block for belonging to a RAID, total storage of a RAID
Capacity of the space size value equal at least about a SSD;If determining that data are the second data according to the attribute of data to be written,
The write of second data is belonged in the memory block of the 2nd RAID, total storage size value of the 2nd RAID is equal to total data disks
Capacity sum.
Further, if the memory block that memory 402 is additionally operable to belong to a RAID is otherwise fully engaged, will be accounted for completely
Belong to the data conversion storage in the memory block of a RAID in the memory block of the 2nd RAID.
Yet further, processor 401 is used for according to the storage size value and the sizes values of memory block of a RAID
It is determined that the memory block number for belonging to a RAID is M, and the storage size value according to the 2nd RAID and memory block is big
Little value determines that the memory block number for belonging to the 2nd RAID is N number of, so that being labeled as belonging to the memory block number of a RAID not
More than M, the memory block number for being labeled as belonging to the 2nd RAID is not more than N.
Wherein, the storage size value of a RAID is according to the determination of the capacity of at least one SSD, the storage of the 2nd RAID
Space size value determines that the sizes values of memory block are a default definite value, and M, N are positive integer according to the capacity of total data disks.
If memory 402 is additionally operable to processor 401 determines that the M memory block for belonging to a RAID is otherwise fully engaged, will
The data belonged in the memory block of a RAID are all dumped in the memory block of the 2nd RAID, and are emptied occupancy M and belonged to the
All data in the memory block of one RAID.
For example, memory 402 is used to determine that the first data need to write the memory block for belonging to a RAID;According to default
The all of memory block of table item index, obtains the address of all memory blocks for being labeled as belonging to a RAID and all unlabelled
The address of memory block;First data are write in corresponding memory block according to the address of memory block.
Such as, memory 402 can be according to all of memory block of default table item index, using unlabelled memory block as first
Memory block, by the memory block not being otherwise fully engaged in all memory blocks for being labeled as and belong to a RAID the second storage is defined as
Block;If there is the second memory block, obtain the address of the second memory block, by the first data according to the second memory block address write the
In two memory blocks;If there is no the second memory block, the address of first memory block is obtained, by the first data according to first memory block
Address write first memory block in, and mark first memory block to belong to a RAID.
Or, memory 402 is used to determine that the second data need to write the memory block for belonging to the 2nd RAID;According to default list item
All of memory block is indexed, the address of all memory blocks for being labeled as belonging to the 2nd RAID and all unlabelled storages is obtained
The address of block;Second data are write in corresponding memory block according to the address of memory block.
Such as, memory 402 can be according to all of memory block of default table item index, using unlabelled memory block as first
Memory block, by the memory block not being otherwise fully engaged in all memory blocks for being labeled as and belong to the 2nd RAID the 3rd storage is defined as
Block;If there is the 3rd memory block, obtain the address of the 3rd memory block, by the second data according to the 3rd memory block address write the
In three memory blocks;If there is no the 3rd memory block, the address of first memory block is obtained, by the second data according to first memory block
Address write first memory block in, and mark first memory block to belong to the 2nd RAID.
The storage device of the present embodiment, can be used for performing the technical scheme of embodiment of the method shown in Fig. 1~Fig. 4, in fact
Existing principle is similar with technique effect, and here is omitted.
Storage device provided in an embodiment of the present invention.Storage device obtains the attribute of data to be written;If according to be written
The attribute of the data for entering determines that data are the first data, then write the first data using the characteristic of solid-state hard disk SSD quick response
In entering the memory block for belonging to a RAID, total storage size value of a RAID is equal at least about the capacity of a SSD;If root
Determine that data are the second data according to the attribute of data to be written, then the write of the second data is belonged to the memory block of the 2nd RAID
In, total storage size value of the 2nd RAID is equal to the capacity sum of total data disks.So, storage device being capable of profit
It is that the data of little input and output IO write to attribute with the fast response characteristic of SSD, to improve the speed of write, and data
Disk only processes the data of big IO, and data disks can directly write the data of the sizes values equal to sector in the data of this kind of big IO
Enter, the speed of write is accelerated with this, reduce because the magnetic head of hard disk frequently reads and writes the change for bringing, lift the service life of hard disk.
Storage device is this to be set while can be accelerated the storage after sector is lifted using the structure and storage mode of SSD and data disk storage
Standby writing speed, and then lift the performance of storage device.
One of ordinary skill in the art will appreciate that:Realizing all or part of step of above-mentioned each method embodiment can lead to
Cross the related hardware of programmed instruction to complete.Aforesaid program can be stored in a computer read/write memory medium.The journey
Sequence upon execution, performs the step of including above-mentioned each method embodiment;And aforesaid storage medium includes:ROM, RAM, magnetic disc or
Person's CD etc. is various can be with the medium of store program codes.
Finally it should be noted that:Various embodiments above only to illustrate technical scheme, rather than a limitation;To the greatest extent
Pipe has been described in detail with reference to foregoing embodiments to the present invention, it will be understood by those within the art that:Its according to
So the technical scheme described in foregoing embodiments can be modified, either which part or all technical characteristic are entered
Row equivalent;And these modifications or replacement, do not make the essence disengaging various embodiments of the present invention technology of appropriate technical solution
The scope of scheme.
Claims (16)
1. a kind of storage device wiring method, it is characterised in that include:
Obtain the attribute of data to be written;
It is quick using solid-state hard disk SSD if determining that the data are the first data according to the attribute of the data to be written
The characteristic of response belongs to first data write in the memory block of the first RAID, and total memory space of a RAID is big
Capacity of the little value equal at least about a SSD;If determining that the data are the second data according to the attribute of the data to be written,
Then second data write is belonged in the memory block of the 2nd RAID, total storage size value of the 2nd RAID is equal to
The capacity sum of total data disks.
2. method according to claim 1, it is characterised in that before the attribute of acquisition data to be written, also wrap
Include:
Determine according to the storage size value of a RAID and the sizes values of the memory block and belong to a RAID
Memory block number be M, and the sizes values of the storage size value according to the 2nd RAID and the memory block determine
The memory block number for belonging to the 2nd RAID is N number of, so that the memory block number for being labeled as belonging to a RAID is not more than
M, the memory block number for being labeled as belonging to the 2nd RAID is not more than N;
Wherein, the storage size value of a RAID according to the capacity of at least one SSD determine, described second
The storage size value of RAID determines that the sizes values of the memory block are one default according to the capacity of total data disks
Definite value, described M, N are positive integer.
3. method according to claim 1, it is characterised in that described that first data write is belonged to into a RAID's
Memory block includes:
Determine that first data need to write the memory block for belonging to a RAID;
According to all of memory block of default table item index, obtain all memory blocks for being labeled as belonging to a RAID address and
The address of all unlabelled memory blocks;
First data are write in corresponding memory block according to the address of the memory block.
4. the method according to any one of claims 1 to 3, it is characterised in that described to belong to second data write
The memory block of the 2nd RAID includes:
Determine that second data need to write the memory block for belonging to the 2nd RAID;
According to all of memory block of default table item index, obtain all memory blocks for being labeled as belonging to the 2nd RAID address and
The address of all unlabelled memory blocks;
Second data are write in corresponding memory block according to the address of the memory block.
5. method according to claim 3, it is characterised in that the basis all of memory block of default table item index, obtains
Take the address of all memory blocks for being labeled as and belonging to a RAID and the address of all unlabelled memory blocks;By described first
Data write corresponding memory block according to the address of the memory block to be included:
According to all of memory block of default table item index, using unlabelled memory block as first memory block, it is labeled as all
The memory block not being otherwise fully engaged in the memory block for belonging to a RAID is defined as the second memory block;
If there is the second memory block, the address of second memory block is obtained, by first data according to the second memory block
Address is write in the second memory block;
If there is no the second memory block, the address of the first memory block is obtained, by first data according to the first storage
In the address write first memory block of block, and the first memory block is marked to belong to a RAID.
6. method according to claim 4, it is characterised in that the basis all of memory block of default table item index, obtains
Take the address of all memory blocks for being labeled as and belonging to the 2nd RAID and the address of all unlabelled memory blocks;By described second
Data write corresponding memory block according to the address of the memory block to be included:
According to all of memory block of default table item index, using unlabelled memory block as first memory block, it is labeled as all
The memory block not being otherwise fully engaged in the memory block for belonging to the 2nd RAID is defined as the 3rd memory block;
If there is the 3rd memory block, the address of the 3rd memory block is obtained, by second data according to the 3rd memory block
Address is write in the 3rd memory block;
If there is no the 3rd memory block, the address of the first memory block is obtained, by second data according to the first storage
In the address write first memory block of block, and the first memory block is marked to belong to the 2nd RAID.
7. the method according to claims 1 to 3,5 any one, it is characterised in that acquisition data to be written take
Before the sizes values in space, also include:
If the memory block for belonging to a RAID is otherwise fully engaged, by the memory block for belonging to a RAID being otherwise fully engaged
In data conversion storage in the memory block of the 2nd RAID.
8. method according to claim 2, it is characterised in that the storage size value according to a RAID
And the sizes values of the memory block determine that the memory block number for belonging to a RAID is M, and according to the 2nd RAID
Storage size value and the memory block sizes values determine belong to the memory block number of the 2nd RAID for it is N number of it
Afterwards, also include:
If the M memory block for belonging to a RAID is otherwise fully engaged, by the data in the memory block for belonging to a RAID
In all dumping to the memory block of the 2nd RAID;
Empty and take the M all data belonged in the memory block of a RAID.
9. a kind of storage device, it is characterised in that include:
Acquiring unit, for obtaining the attribute of data to be written;
Write control unit, if for determining the data for the first data, profit according to the attribute of the data to be written
First data write is belonged in the memory block of a RAID with the characteristic of solid-state hard disk SSD quick response, described first
Capacity of the total storage size value of RAID equal at least about a SSD;If being determined according to the attribute of the data to be written
The data are the second data, then belong to second data write in the memory block of the 2nd RAID, the 2nd RAID's
Total storage size value is equal to the capacity sum of total data disks.
10. storage device according to claim 9, it is characterised in that also include:
Processing unit, belongs to for determining according to the sizes values of the storage size value of a RAID and the memory block
The memory block number of the first RAID is M, and according to the storage size value and the memory block of the 2nd RAID
Sizes values determine and belong to the memory block number of the 2nd RAID for N number of so that being labeled as belonging to the storage of a RAID
Block number is not more than M, and the memory block number for being labeled as belonging to the 2nd RAID is not more than N;
Wherein, the storage size value of a RAID according to the capacity of at least one SSD determine, described second
The storage size value of RAID determines that the sizes values of the memory block are one default according to the capacity of total data disks
Definite value, described M, N are positive integer.
11. storage devices according to claim 9, it is characterised in that said write control unit includes:
Determining module, for determining that first data need to write the memory block for belonging to a RAID;
Acquisition module, for according to all of memory block of default table item index, acquisition is all to be labeled as belonging to depositing for a RAID
The address of storage block and the address of all unlabelled memory blocks;
Writing module, for first data to be write in corresponding memory block according to the address of the memory block.
12. storage devices according to any one of claim 9~11, it is characterised in that
The determining module, is additionally operable to determine that second data need to write the memory block for belonging to the 2nd RAID;
The acquisition module, is additionally operable to according to all of memory block of default table item index, and acquisition is all to be labeled as belonging to second
The address of the memory block of RAID and the address of all unlabelled memory blocks;
Said write module, is additionally operable to second data be write in corresponding memory block according to the address of the memory block.
13. storage devices according to claim 11, it is characterised in that
The determining module, specifically for according to the default all of memory block of table item index, using unlabelled memory block as the
One memory block, by the memory block not being otherwise fully engaged in all memory blocks for being labeled as and belong to a RAID the second storage is defined as
Block;
The acquisition module, if specifically for there is the second memory block, obtaining the address of second memory block, said write mould
Block, specifically for first data are write in the second memory block according to the address of the second memory block;
The acquisition module, it is described to write if specifically for there is no the second memory block, obtaining the address of the first memory block
Enter module, specifically for first data are write in first memory block according to the address of first memory block, and mark described
First memory block belongs to a RAID.
14. storage devices according to claim 12, it is characterised in that
The determining module, specifically for according to the default all of memory block of table item index, using unlabelled memory block as the
One memory block, by the memory block not being otherwise fully engaged in all memory blocks for being labeled as and belong to the 2nd RAID the 3rd storage is defined as
Block;
The acquisition module, if specifically for there is the 3rd memory block, obtaining the address of the 3rd memory block, said write mould
Block, specifically for second data are write in the 3rd memory block according to the address of the 3rd memory block;
The acquisition module, it is described to write if specifically for there is no the 3rd memory block, obtaining the address of the first memory block
Enter module, second data are write in first memory block according to the address of first memory block, and mark first storage
Block belongs to the 2nd RAID.
15. storage devices according to claim 9~11,13 any one, it is characterised in that
Said write control unit, if the memory block for being additionally operable to belong to a RAID is otherwise fully engaged, is accounted for described completely
Belong to the data conversion storage in the memory block of a RAID in the memory block of the 2nd RAID.
16. storage devices according to claim 10, it is characterised in that
Said write control unit, if the M memory block for belonging to a RAID for being additionally operable to the processing unit determination is complete
Take, then all dump to the data in the memory block for belonging to a RAID in the memory block of the 2nd RAID, and empty
Take all data in the M memory block for belonging to a RAID.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310724721.2A CN103714010B (en) | 2013-12-24 | 2013-12-24 | Storage device write-in method and storage device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310724721.2A CN103714010B (en) | 2013-12-24 | 2013-12-24 | Storage device write-in method and storage device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103714010A CN103714010A (en) | 2014-04-09 |
CN103714010B true CN103714010B (en) | 2017-04-12 |
Family
ID=50407007
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310724721.2A Active CN103714010B (en) | 2013-12-24 | 2013-12-24 | Storage device write-in method and storage device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103714010B (en) |
Families Citing this family (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106033320B (en) * | 2015-03-11 | 2020-04-14 | 中兴通讯股份有限公司 | Processing method and device for high-performance Redundant Array of Inexpensive Disks (RAID) mechanism |
CN107870731B (en) * | 2016-09-23 | 2021-07-27 | 伊姆西Ip控股有限责任公司 | Management method of Redundant Array of Independent Disks (RAID) system and electronic equipment |
CN108008912A (en) * | 2016-11-02 | 2018-05-08 | 华为技术有限公司 | A kind of memory buffers method and intelligent terminal based on DRAM |
CN107329694B (en) * | 2017-06-22 | 2021-05-18 | 苏州奈特力智能科技有限公司 | Data storage control method, control device and storage equipment |
CN107967121B (en) * | 2017-10-25 | 2020-04-14 | 华为技术有限公司 | Data writing method and storage device |
CN113282582B (en) * | 2021-05-21 | 2023-06-20 | 海南超船电子商务有限公司 | Efficient storage method and system for ship position data |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101727402A (en) * | 2009-10-23 | 2010-06-09 | 深圳市江波龙电子有限公司 | Read and write control method and system of data of nonvolatile storage |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7194596B2 (en) * | 2004-06-09 | 2007-03-20 | Simpletech Global Limited | Method of efficient data management with flash storage system |
-
2013
- 2013-12-24 CN CN201310724721.2A patent/CN103714010B/en active Active
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101727402A (en) * | 2009-10-23 | 2010-06-09 | 深圳市江波龙电子有限公司 | Read and write control method and system of data of nonvolatile storage |
Also Published As
Publication number | Publication date |
---|---|
CN103714010A (en) | 2014-04-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103714010B (en) | Storage device write-in method and storage device | |
US9292218B2 (en) | Method and apparatus to manage object based tier | |
US20180024921A1 (en) | Memory system executing garbage collection | |
CN102792284B (en) | Memory devices loss equalization technology | |
US8677093B2 (en) | Method and apparatus to manage tier information | |
US10360144B2 (en) | Storage apparatus and non-volatile memory device including a controller to selectively compress data based on an update frequency level | |
US10162554B2 (en) | System and method for controlling a programmable deduplication ratio for a memory system | |
TWI438778B (en) | Method for suppressing errors, and associated memory device and controller thereof | |
JP2012519899A (en) | Select memory block | |
US20170038977A1 (en) | Valid data merging method, memory controller and memory storage apparatus | |
US11809723B2 (en) | Unbalanced plane management method, associated data storage device and controller thereof | |
CN101714065A (en) | Method for managing mapping information of flash controller | |
CN107291392A (en) | A kind of solid state hard disc and its reading/writing method | |
CN101604227A (en) | The method and apparatus of data storage | |
TWI388986B (en) | Flash memory apparatus and method for operating a flash memory apparatus | |
CN104484135A (en) | Method and device for quickly reading data | |
CN107229580A (en) | Sequential stream detection method and apparatus | |
JP2009259145A (en) | Memory controller, flash memory system provided with memory controller, and method for controlling flash memory | |
TWI752784B (en) | Unbalanced plane management method, associated data storage device and controller thereof | |
CN105138294B (en) | A kind of hard disc apparatus and information processing method | |
TWI453747B (en) | Method for managing a plurality of blocks of a flash memory, and associated memory device and controller thereof | |
CN105260261B (en) | A kind of mail restoration methods | |
TWI763050B (en) | Self-adaptive wear leveling method and algorithm and related memory device and apparatus | |
CN109669620A (en) | Storage management method, memorizer control circuit unit and memorizer memory devices | |
CN105260882B (en) | A kind of mail high-efficient treatment 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 | ||
TR01 | Transfer of patent right |
Effective date of registration: 20220907 Address after: No. 1899 Xiyuan Avenue, high tech Zone (West District), Chengdu, Sichuan 610041 Patentee after: Chengdu Huawei Technologies Co.,Ltd. Address before: 518129 Bantian HUAWEI headquarters office building, Longgang District, Guangdong, Shenzhen Patentee before: HUAWEI TECHNOLOGIES Co.,Ltd. |
|
TR01 | Transfer of patent right |