CN107229535A - Many copy storage methods of data block, storage device, method for reading data - Google Patents
Many copy storage methods of data block, storage device, method for reading data Download PDFInfo
- Publication number
- CN107229535A CN107229535A CN201710367522.9A CN201710367522A CN107229535A CN 107229535 A CN107229535 A CN 107229535A CN 201710367522 A CN201710367522 A CN 201710367522A CN 107229535 A CN107229535 A CN 107229535A
- Authority
- CN
- China
- Prior art keywords
- copy
- data block
- data
- storage device
- read
- 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
- G06F11/00—Error detection; Error correction; Monitoring
- G06F11/07—Responding to the occurrence of a fault, e.g. fault tolerance
- G06F11/14—Error detection or correction of the data by redundancy in operation
- G06F11/1402—Saving, restoring, recovering or retrying
- G06F11/1446—Point-in-time backing up or restoration of persistent data
- G06F11/1448—Management of the data involved in backup or backup restore
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application provides a kind of many copy storage methods of data block, storage device, method for reading data.Wherein, during many copy storages, storage device obtains specific field from the SCSI write command of reception, the copy of the quantity writing data blocks of the copy indicated according to specific field.Realize and many copies formed in units of data block, it is ensured that the high reliability of significant data, at the same avoid whole LUN is backed up, the operation such as mirror image, dramatically saves on memory space.In read block, storage device read block first, if reading failure, judges whether the data block has copy, if with copy, the copy of the data block is read at the position of copy.Realize in the case where data block reads failure, directly utilize storage device automatic detection and read copy, the maximum duration of whole data read process is a millisecond magnitude, fundamentally solves and causes the problem of business continuance is damaged using repairing software and repair the operation such as file in the prior art.
Description
Technical field
The application is related to technical field of data storage, more particularly to many copy storage methods of data block a kind of, storage are set
Standby, method for reading data.
Background technology
Data safety is one of important subjects of subject such as computer and network, and it is not only related to personal, enterprise
Industry, it is often more important that it may directly affect national security, therefore kept away for important data, it is necessary to carry out the operation such as backing up
Exempt to cause loss of data because bottom storage is abnormal, for example, avoid due to RAID (Redundant Arrays of
Independent Disks, RAID) failure loss data.
For above-mentioned scene demand, there are a variety of solutions in the prior art, for example, data backup, data image, double
Living and remote copy etc..But be that current significant data and insignificant data are normally at same LUN the problem of exist
On (Logical Unit Number, logic unit), it is difficult to be separated, therefore it can only carry out the behaviour such as backing up for whole LUN
Make, then many insignificant data such as can also be backed up together at the operation along band, may result in the serious waste of memory space.
The problem of to solve waste of storage space in the operation such as above-mentioned backup, propose in the prior art a kind of for file-level
The processing scheme of other data, i.e., carry out the operation such as backing up using specific software for the important file in part.This method can
Largely to save memory space, but be the problem of exist if file corruption, it is necessary to by corresponding software come
Repaired, it is necessary to after pending file is repaired from Backup Data, business could be recovered, when the repair process needs certain
Between, the continuity for business affects.
The content of the invention
In view of this, the application provides a kind of many copy storage methods of data block, storage device, method for reading data,
On the premise of data high reliability is ensured, solve data backup in the prior art etc. operation caused by storage resource waste with
And the impaired technical problem of the business continuance that causes of file reparation.
Specifically, the application is achieved by the following technical solution：
In a first aspect, the embodiment of the present application provides a kind of many copy storage methods of data block, applied to storage device,
Methods described comprises the following steps：
Receive the SCSI write command for performing write operation to data block；
Specific field is obtained from the SCSI write command, the specific field is used for the copy for indicating the data block
Quantity；
The quantity indicated according to the specific field writes the copy of the data block.
Alternatively, the copy that the quantity indicated according to the specific field writes the data block includes following step
Suddenly：
Determine the RAID of the data block write-in；
From the RAID in addition to the RAID selection with the quantity identical RAID, by the copy be respectively written into
Each RAID of selection.
Alternatively, methods described also comprises the following steps：
Copy mark is set in the corresponding metadata of the data block, and utilizes copy described in the metadata record
Position.
Alternatively, the copy amount for the data block that the specific field is indicated is the outside by being connected with the storage device
Server is set according to the copy command that user sends, and the copy amount of data block is included in the copy command.
Alternatively, the copy amount for the data block that the specific field is indicated is the outside by being connected with the storage device
Server is set according to the copy amount obtained from local data input port.
Second aspect, the embodiment of the present application provides a kind of storage device, and the equipment includes：
Order receiving unit, for receiving the SCSI write command for being used for that write operation to be performed to data block；
Field acquiring unit, for obtaining specific field from the SCSI write command, the specific field is used to indicate
The quantity of the copy of the data block；
Data write unit, the quantity for being indicated according to the specific field writes the copy of the data block.
Alternatively, the data write unit includes：
Data block RAID determination subelements, the RAID for determining the data block write-in；
Subelement is write, for the selection from the RAID in addition to the RAID and the quantity identical RAID, and will
The copy is respectively written into each RAID of selection.
Alternatively, the storage device also includes metadata record mark unit, and the metadata record mark unit is used
In setting copy mark in the corresponding metadata of the data block, and utilize the position of copy described in the metadata record.
The third aspect, the embodiment of the present application provides a kind of method for reading data, and methods described comprises the following steps：
Receive data block and read instruction；
The position that instruction determines the data block to be read is read according to the data block；
The data block is read since the position of the data block；If reading failure, whether the data block is judged
With copy mark；Wherein, the copy mark is used to indicate that the data block has copy；
If the data block has the copy mark, the position of the copy of the data block is searched, and according to described
Read the copy of the data block in the position of copy.
Fourth aspect, the embodiment of the present application provides a kind of storage device, and the storage device includes：
Instruction reception unit, instruction is read for receiving data block；
Position determination unit, for reading the position that instruction determines the data block to be read according to the data block；
Data-reading unit, for reading the data block since the position of the data block；
If reading failure, the data-reading unit is additionally operable to, and judges whether the data block has copy mark；If
The data block has the copy mark, then searches the position of the copy of the data block, and according to the position of the copy
Read the copy of the data block；Wherein, the copy mark is used to indicate that the data block has copy.
As shown from the above technical solution, the embodiment of the present application is in data storage, and storage device writes life from the SCSI of reception
Specific field, and the copy of the quantity writing data blocks of the copy indicated according to specific field are obtained in order.The application can be with
Data block be unit form many copies, it is ensured that the high reliability of significant data, at the same avoid whole LUN is backed up,
Mirror image etc. is operated, and dramatically saves on the memory space of storage device.
The present patent application embodiment is in read block, and storage device reads instruction from the position of data block according to data block
Put beginning read block.If reading failure, judge whether the data block has copy, if the data block has copy,
The copy of the data block is read at the position of copy.The application directly can be set when data block reads failure using storage
Standby automatic detection copy simultaneously reads copy, and the maximum duration of whole data read process is a millisecond magnitude, it is entirely avoided existing
Have in technology using the defect that software reparation file etc. operates the business continuance caused to be damaged is repaired, substantially increase copy reading
The fluency taken, improves user experience.
Brief description of the drawings
Fig. 1 is a kind of many copy storage method FB(flow block)s of data block shown in the exemplary embodiment of the application one；
Fig. 2 is many copy storage method FB(flow block)s of another data block shown in the exemplary embodiment of the application one；
Fig. 3 is a kind of method for reading data FB(flow block) shown in the exemplary embodiment of the application one；
Fig. 4 is another method for reading data FB(flow block) shown in the exemplary embodiment of the application one；
Fig. 5 is a kind of storage device block diagram shown in the exemplary embodiment of the application one；
Fig. 6 is another storage device block diagram shown in the exemplary embodiment of the application one；
Fig. 7~Figure 11 is to specify location position view in the embodiment of the present application in SCSI write command.
Embodiment
Here exemplary embodiment will be illustrated in detail, its example is illustrated in the accompanying drawings.Following description is related to
During accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawings represent same or analogous key element.Following exemplary embodiment
Described in embodiment do not represent all embodiments consistent with the application.On the contrary, they be only with it is such as appended
The example of the consistent apparatus and method of some aspects be described in detail in claims, the application.
It is the purpose only merely for description specific embodiment in term used in this application, and is not intended to be limiting the application.
" one kind ", " described " and "the" of singulative used in the application and appended claims are also intended to including majority
Form, unless context clearly shows that other implications.It is also understood that the term "and/or" that may be used herein is
Refer to and may be combined comprising one or more associated any or all of project listed.
It will be appreciated that though various information, but this may be described using term first, second, third, etc. in the application
A little information should not necessarily be limited by these terms.These terms are only used for same type of information being distinguished from each other out.For example, not departing from
In the case of the application scope, the first information can also be referred to as the second information, similarly, and the second information can also be referred to as
One information.Depending on linguistic context, word as used in this " if " can be construed to " ... when " or " when ...
When " or " in response to determining ".
For convenience of understanding, the application is first illustrated to the noun that may hereinafter use.
RAID：Redundant Arrays of Independent Disks, RAID, it is by very
The relatively inexpensive disk of many prices, is combined into the huge disk group of a capacity, addition produced by providing data using indivedual disks
The whole disk system efficiency of effect promoting.
SCSI:Small Computer System Interface, are that life is transmitted between a kind of main frame and storage device
Make, state and block number according to etc. traffic operation specification set.
SBC：SCSI Block Commands, SCSI block device instruction set.
Metadata：Metadata, a kind of data of record management relevant information, it is understood that it is that one kind is used for describing to be
The data of data, metadata is stored on a storage device, for example, LUN metadata is stored on RAID.
To solve the technical problem for operating the storage resource caused to waste by backup, mirror image etc. in the prior art, the application
Embodiment provides a kind of many copy storage methods of data block, applied to storage device, as shown in figure 1, this method is included such as
Lower step：
S11, receive for data block perform write operation SCSI write command；
In step S11, a data block can be formed by performing a SCSI write command；
S12, obtain specific field from SCSI write command, specific field is used for the quantity for indicating the copy of data block；
Specifically, it is possible to use the pre-determined bit of the corresponding byte of the specific field indicates the quantity of copy, wherein predetermined
Binary data is set in position, the quantity of copy is indicated using the binary data of setting.For example utilize the specific field pair
The byte answered the 5th, 6 quantity to define the copy for needing to write, if the 5th of the byte the, 6 be set to 01 expression should
Data block needs the quantity of the copy write to be 1, if the 5th of the byte the, 6 be set to the 10 expressions data blocks and need to write
The quantity of the copy entered is 2, if the 5th of the byte the, 6 be set to the number of copy that the 11 expressions data blocks needs to write
Measure as 3.Certain corresponding byte of the specific field the 5th, 6 may be arranged as 00, represent that the data block need not be write
Enter copy.The specific number of copy can flexibly be set according to the significance level of the data block in different application scenarios,
The high reliability of significant data is ensure that, while effectively reducing the waste of storage resource.In addition, the application is not limited to two
The quantity of copy is indicated as pre-determined bit, can for example may be used with the quantity that copy is indicated as pre-determined bit such as 3 or 4
Using with the 4th of the corresponding byte of the specific field, 5,6 as pre-determined bit, to define the quantity for the copy for needing to write.
S13, the quantity writing data blocks indicated according to specific field copy.
Those skilled in the art it is appreciated that the various the application write one or more copies while, before or after
Also it is written with data block in itself.
Many copy storage methods of above-mentioned data block, have simply carried out a bit to the write instructions in existing SCSI protocol
Improve, i.e., the quantity of copy is provided with the reserved field of SCSI write commands, can just deposited when performing SCSI write commands
Multiple copies that data block is formed in equipment are stored up, it is simple and convenient, the security of data block has been effectively ensured, simultaneously because this
Application is the copy formed in units of data block, therefore effectively save the memory space of storage device, it is to avoid storage money
The waste in source.
It should be noted that in above-mentioned steps S12, specific field is any undefined in preassigned SCSI write command
Field, in one embodiment, during above-mentioned specific field can be the reserved field in SCSI write command, i.e. Fig. 7~Figure 11
The Reserved fields shown.Specify the quantity of copy not produce former SCSI write command using reserved field Reserved to appoint
What influences.Each write command in SBC agreements, i.e., each SCSI write command can have several untapped reserved fields, such as Fig. 7
Shown in~Figure 11, the position that location is retained in each SCSI write command is different, can be according to not during storage device parsing SCSI protocol
Same write command is then parsed pre- in the correlation reserved field of each SCSI write command in corresponding location lookup reserved field
Position is stayed, the quantity of copy is determined.
One or more copies of data block are formd by performing step S13.It should be noted that each copy difference
In a RAID, i.e., different copies are located in different RAID, so wherein a RAID failure or when damaging it is unlikely
In the safety for the copy for influenceing to store in other RAID, that is, the RAID for realizing the copy that is stored with wherein breaks down
When, it can be stored with from other in RAID of copy and read copy.Likewise, the RAID for data storage block is with storing each
The RAID of copy is differed, and being so unlikely to influence in RAID failure or the damage of data storage block is stored in other RAID
In copy safety, that is, when realizing the RAID failure of data storage block, can be stored with from other in RAID of copy read
Copy.The method of above-mentioned data storage block and copy in different RAID is effectively improved the data for the significant data that is stored with
The security of block.Realize that above-mentioned write operation specifically comprises the following steps：
S131, the RAID for determining data block write-in；
S132, selection and the quantity identical RAID of copy from the RAID in addition to RAID, by copy be respectively written into
Each RAID of selection.
It should be noted that the copy of each in the application may be located in different disk chassis or different storages set
It is standby upper.
One or more copies of the data block formed after above-mentioned steps S11~S13 in storage device are performed, but only
Store, and it is skimble-skamble that can not read out, to realize the reading to the copy of storage, the above method also needs to include
Following steps：
S14, in LUN metadata, copy mark is specifically set in the corresponding metadata of data block, and utilizes member
The position of data record copy.
In step S14, copy mark is set to be used to indicate which data block has copy, in read block failure
When, first determine whether whether the data block has copy mark, that is, first determine whether whether the data block has copy.It is determined that having
After copy, the position of the copy of the data block is searched in the metadata for recording the position of copy of each data block, for example,
The skew of uuid (being used to identify RAID where copy) and copy in the RAID of RAID where searching copy (is used to identify pair
Position of the sheet in the RAID), the position of copy is assured that according to this 2 information.Afterwards can according to the position of copy
Copy is read from storage device.
In the corresponding metadata of data block behind the position of transcript mark and copy, the mistake of copy is being read below
Journey, storage device can be automatically performed, and the time for reading copy is a millisecond magnitude, it is entirely avoided due to prior art Chinese
The service disconnection that the operation such as part reparation is caused.
It should be noted that the SCSI write command of above-mentioned execution data block write operation is taken by the outside being connected with storage device
The generation of business device and transmission.The copy amount in specific field in SCSI write command is to be sent by external server according to user
Copy command is set.In one embodiment, the external server sets the number of copies in specific field by following step
Amount：
S31, external server receive copy command；
Wherein the copy command is a customized special copy command in file system, and user or application layer pass through
The copy command can set the quantity for the copy for needing to write, and such as copy command is copy2, then represent file destination
In all data blocks need the quantity of the copy generated to be 2 parts.
S32, copy command parse obtain the quantity of copy, and specific field is set according to the quantity of copy.Tool
Body, i.e., can be generation mesh when SCSI command encapsulated layer encapsulates SCSI command when calling bottom to encapsulate the module of SCSI protocol
Mark the quantity that the specific field in all SCSI write commands of file inserts the copy that parsing is obtained.
It will be clear for those skilled in the art that from the angle of file system, file is minimum storage cell, text
When any one data block that part is included is abnormal, this document all can not normally be read.Even if therefore only one of which in this document
Data block it is important, to ensure that the important data block can normally be read, other data blocks of this document are also allowed for just
Often read, then all data blocks of this document are required to stored copies.Above-mentioned first server will be taken in predetermined copy command
The quantity of the copy of band, has inserted the specific field for all SCSI write commands for generating the file destination, then the file destination
All data blocks there is the copy of identical quantity, the safety of file destination has been fully ensured that, so as to ensure that in file destination
The safety of critical data blocks.
It should be noted that designated word in the SCSI write command of above-mentioned execution data block write operation in another embodiment
The copy amount for the data block that section is indicated can also be inputted by the external server being connected with storage device according to from local data
The copy amount that port is obtained is set.Comprise the following steps that：
S41, outside business device obtain predetermined flag information；
The predetermined flag information can be customized when file destination is opened, and the predetermined flag can be transmitted
The module of SCSI protocol is encapsulated to SCSI command encapsulated layer, i.e. bottom, all of execution required for generating the file destination are represented
The specific field of SCSI write command is required to insert copy amount.
S42, from data-in port obtain copy amount, and according to the quantity of copy set generate the file destination needed for
Specific field in all SCSI write commands to be performed.In this step, application layer is set in data-in port needs write-in
Copy quantity.
The specific field that all SCSI write commands of file destination will also be generated in the above-described embodiments has inserted identical
Quantity, that is, all data blocks of the file destination generated all have the copy of identical quantity.In above-mentioned steps S31 in copy command
The quantity of the copy obtained in the quantity and step S41 of the copy of carrying by data-in port is all application layer according to reality
Flexibly set the need for border.As shown in Fig. 2 after the quantity of acquisition copy, SCSI command encapsulated layer exists according to above-mentioned quantity
The specific field of each SCSI write commands inserts corresponding copy amount, so far completes the improvement of SCSI write commands.
Above-mentioned utilization external server sets the embodiment of specific field to encapsulate SCSI command just with file system
A kind of scene, certainly can also be using other means encapsulation SCSI command, the application being for example written and read directly against block device,
These applications just can directly encapsulate SCSI command, and be handed down to storage device.
It should be noted that as shown in Fig. 2 to judge whether to be successfully written data block and/or its copy, in write-in
Data block and/or its copy need to return to write successfully identify after completing, and specifically include two kinds of scenes, the first scene
It is, when data block does not need stored copies, after writing data blocks are completed, you can return to a write-in successfully mark.The
Two kinds of scenes are, when data block needs stored copies, it is necessary to after data block and all copies all write successfully, ability
It is enough to return to write-in successfully mark.
As shown in Fig. 2 the quantity of the copy write the need for application layer setting is also likely to be 0, then avoid the need for write-in
Copy, then SCSI write commands, which are normally written after data block, returns to write-in successfully mark.
As shown in Fig. 2 when the quantity of the copy of write-in is not 0, the RAID of write-in copy must be available RAID, i.e.,
The remaining space of the RAID can accommodate copy.
On a storage device in the concrete operations of write-in copy, the minimum memory list using the data block of fixed length as copy
Position, the data block of such as fixed length is 4KB.
In summary, the embodiment of above-mentioned many copy storage methods can be generated many in units of data block to significant data
Copy, it is ensured that the safety of significant data, while saving the storage resource of preciousness.Simultaneously because storing pair in storage device
The position of this mark and copy so that storage device corresponding with this method can be automatically performed the reading of copy, improve
In the situation of data block read failure, the continuity of related reading business improves user experience.
Embodiment with foregoing many copy storage methods is corresponding, and present invention also provides a kind of implementation of storage device
Example.As shown in figure 5, the equipment includes：
Order receiving unit, for receiving the SCSI write command for being used for that write operation to be performed to data block；
Field acquiring unit, for obtaining specific field from SCSI write command, specific field is used to indicate data block
The quantity of copy；
Data write unit, for the copy of the quantity writing data blocks indicated according to specific field.
It should be noted that data write unit includes：
Data block RAID determination subelements, the RAID for determining data block write-in；
Subelement is write, for the selection from the RAID in addition to RAID and quantity identical RAID, and copy is distinguished
Each RAID of write-in extremely selection.
It should be noted that storage device, which also includes metadata record, identifies unit, metadata record mark unit is used for
Copy mark is set in the corresponding metadata of data block, and utilizes the position of metadata record copy.
The function of unit or subelement and the implementation process of effect specifically refer to above-mentioned more secondary in above-mentioned storage device
The implementation process of correspondence step, will not be repeated here in this storage method.
Many copy storage methods are above mentioned embodiment provided, the storage to copy is realized, example below corresponds to upper
The copy for the storage stated is there is provided a kind of method for reading data, as shown in figure 3, this method comprises the following steps：
S21, reception data block read instruction；
It is to be handed down to storage device by front end applications that data block in this step, which reads instruction, for read block；
S22, the position according to the data block reading instruction determination data block to be read；
S23, the read block since the position of data block；If reading failure, judge whether data block has copy
Mark；Wherein, copy mark is used to indicate that data block has copy；
If data block has copy mark, according to the position of the copy of copy mark searching data block, and according to copy
Position read block copy；
In this step S23, the position of copy mark and copy is specifically searched in the metadata.
It should be noted that in step S22, reading instruction according to data block and determining that the position for the data block to be read has
Body includes two kinds of scenes, and the first scene is that the position of data block is directly carried on data block and read in instruction, should by parsing
Data block, which reads instruction, just can directly determine the position of data block.Second of scene is that the position of data block is not carried on number
Read according to block in instruction, data block goes to read the mark that the data block is only carried in instruction, it is necessary to further basis
The mark determines the position of the data block.
It should be noted that in step S23, two kinds of scenes are included according to the copy of the position read block of copy, the
A kind of scene is that copy is read successfully, now returns to the copy of reading；Second of scene is that all copies read and lost
Lose, now the reading utter failure of the data block is, it is necessary to which returned data reads the information of failure.
It should be noted that the step of reading copy when data block reads failure is described in detail in above-mentioned steps S23, but
Be exist in actual reading process data block read successfully scene, if as shown in figure 4, data block is read successfully, only needing
Return to the data read from data block, it is not necessary to determine whether whether the data block has copy mark.
Data block is described in detail in above-mentioned steps S23 and reads failure, and there is the data block processing during copy to walk
Suddenly, but in actual reading process there is the scene that the data block does not have copy, as shown in figure 4, the now data block
Digital independent utter failure, it is necessary to returned data read failure information.
Many method for reading data that above-described embodiment is provided realize storage device and automatic jump to reading at locations of copies
The function of copy, the whole time for reading data is a millisecond magnitude, and the continuity that client reads data will not be impacted,
User experience is improved, while ensure that the high reliability of data.
Embodiment with aforementioned data read method is corresponding, and present invention also provides the implementation of another storage device
Example.As shown in fig. 6, the equipment includes：
Instruction reception unit, instruction is read for receiving data block；
Position determination unit, for reading the position that instruction determines the data block to be read according to data block；
Data-reading unit, for the read block since the position of data block；
If reading failure, data-reading unit is additionally operable to judge whether data block has copy mark；If data block has
The position of copy mark, the then copy of searching data block, and the copy of the position read block according to copy；Wherein, copy
Identify for indicating that data block has copy.
The function of unit and the implementation process of effect specifically refer to above-mentioned method for reading data in above-mentioned storage device
The implementation process of middle correspondence step, will not be repeated here.
The preferred embodiment of the application is the foregoing is only, not to limit the application, all essences in the application
God is with principle, and any modification, equivalent substitution and improvements done etc. should be included within the scope of the application protection.
Claims (10)
1. many copy storage methods of a kind of data block, it is characterised in that methods described is applied to storage device, including following step
Suddenly：
Receive the SCSI write command for performing write operation to data block；
Specific field is obtained from the SCSI write command, the specific field is used for the number for indicating the copy of the data block
Amount；
The quantity indicated according to the specific field writes the copy of the data block.
2. many copy storage methods according to claim 1, it is characterised in that described to be indicated according to the specific field
The copy that quantity writes the data block comprises the following steps：
Determine the RAID of the data block write-in；
Selection and the quantity identical RAID, the copy is respectively written into selection from the RAID in addition to the RAID
Each RAID.
3. many copy storage methods according to claim 1 or 2, it is characterised in that methods described also comprises the following steps：
Copy mark is set in the corresponding metadata of the data block, and utilizes the position of copy described in the metadata record
Put.
4. many copy storage methods according to claim 1, it is characterised in that the data block that the specific field is indicated
Copy amount is to be set by the external server being connected with the storage device according to the copy command that user sends, the copy
The copy amount of data block is included in order.
5. many copy storage methods according to claim 1, it is characterised in that the data block that the specific field is indicated
Copy amount is according to the number of copies obtained from local data input port by the external server being connected with the storage device
Amount is set.
6. a kind of storage device, it is characterised in that the equipment includes：
Order receiving unit, for receiving the SCSI write command for being used for that write operation to be performed to data block；
Field acquiring unit, for obtaining specific field from the SCSI write command, the specific field is described for indicating
The quantity of the copy of data block；
Data write unit, the quantity for being indicated according to the specific field writes the copy of the data block.
7. storage device according to claim 6, it is characterised in that the data write unit includes：
Data block RAID determination subelements, the RAID for determining the data block write-in；
Subelement is write, for the selection from the RAID in addition to the RAID and the quantity identical RAID, and will be described
Copy is respectively written into each RAID of selection.
8. the storage device according to claim 6 or 7, it is characterised in that the storage device also includes metadata record
Unit is identified, the metadata record mark unit is used to set copy mark in the corresponding metadata of the data block, and
Utilize the position of copy described in the metadata record.
9. a kind of method for reading data, it is characterised in that methods described comprises the following steps：
Receive data block and read instruction；
The position that instruction determines the data block to be read is read according to the data block；
The data block is read since the position of the data block；If reading failure, judge whether the data block has
Copy mark；Wherein, the copy mark is used to indicate that the data block has copy；
If the data block has the copy mark, the position of the copy of the data block is searched, and according to the copy
Position read the copy of the data block.
10. a kind of storage device, it is characterised in that the storage device includes：
Instruction reception unit, instruction is read for receiving data block；
Position determination unit, for reading the position that instruction determines the data block to be read according to the data block；
Data-reading unit, for reading the data block since the position of the data block；
If reading failure, the data-reading unit is additionally operable to, and judges whether the data block has copy mark；If described
Data block has the copy mark, then searches the position of the copy of the data block, and read according to the position of the copy
The copy of the data block；Wherein, the copy mark is used to indicate that the data block has copy.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710367522.9A CN107229535B (en) | 2017-05-23 | 2017-05-23 | Multi-copy storage method, storage device and data reading method for data block |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710367522.9A CN107229535B (en) | 2017-05-23 | 2017-05-23 | Multi-copy storage method, storage device and data reading method for data block |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107229535A true CN107229535A (en) | 2017-10-03 |
| CN107229535B CN107229535B (en) | 2020-01-21 |
Family
ID=59933816
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710367522.9A Active CN107229535B (en) | 2017-05-23 | 2017-05-23 | Multi-copy storage method, storage device and data reading method for data block |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107229535B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110032333A (en) * | 2017-11-29 | 2019-07-19 | 三星电子株式会社 | Storage system and its operating method |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8775717B2 (en) * | 2007-12-27 | 2014-07-08 | Sandisk Enterprise Ip Llc | Storage controller for flash memory including a crossbar switch connecting a plurality of processors with a plurality of internal memories |
| CN103970484A (en) * | 2013-02-06 | 2014-08-06 | 华为技术有限公司 | Method, device and system for controlling data storage |
| CN103699494B (en) * | 2013-12-06 | 2017-03-15 | 北京奇虎科技有限公司 | A kind of date storage method, data storage device and distributed memory system |
| CN105577776A (en) * | 2015-12-17 | 2016-05-11 | 上海爱数信息技术股份有限公司 | Distributed storage system and method based on data arbiter copy |
| CN105872040B (en) * | 2016-03-30 | 2018-11-30 | 华中科技大学 | A method of write performance is stored using gateway node cache optimization distributed block |
-
2017
- 2017-05-23 CN CN201710367522.9A patent/CN107229535B/en active Active
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110032333A (en) * | 2017-11-29 | 2019-07-19 | 三星电子株式会社 | Storage system and its operating method |
| CN110032333B (en) * | 2017-11-29 | 2024-05-31 | 三星电子株式会社 | Memory system and method of operating the same |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107229535B (en) | 2020-01-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10705932B2 (en) | Method, device and computer program product for managing a storage system | |
| CN102024044B (en) | Distributed file system | |
| US7103811B2 (en) | Mechanisms for detecting silent errors in streaming media devices | |
| US11003536B2 (en) | Method, device and computer readable storage medium for writing to disk array | |
| US20130254457A1 (en) | Methods and structure for rapid offloading of cached data in a volatile cache memory of a storage controller to a nonvolatile memory | |
| CN105183581A (en) | Database disaster tolerance system | |
| US20170091052A1 (en) | Method and apparatus for redundant array of independent disks | |
| WO2014101412A1 (en) | Disk reconstruction method and device therefor | |
| KR20060043873A (en) | System and method for drive recovery following a drive failure | |
| US20180357142A1 (en) | Reporting defects in a flash memory back-up system | |
| CN101815983A (en) | Method and system for preventing corruption of hard disk drive file system | |
| CN102187311A (en) | Methods and systems for recovering a computer system using a storage area network | |
| CN100437504C (en) | Hard disc data backup system and method | |
| CN107870731A (en) | The management method and electronic equipment of redundant array of independent disks system | |
| WO2024113685A1 (en) | Data recovery method for raid array and related apparatus | |
| CN108205482B (en) | File mount restoration methods | |
| JP2005122453A (en) | Disk controller control system for storage device, and the storage device | |
| CN108234465B (en) | Redundancy method and device for handling exception in distributed file system | |
| US7600151B2 (en) | RAID capacity expansion interruption recovery handling method and system | |
| US20050091266A1 (en) | Data file system, data access server and data access program storage medium | |
| US20070043968A1 (en) | Disk array rebuild disruption resumption handling method and system | |
| CN102495815B (en) | Processing method for I/O (input/output) data access interrupt, system and equipment | |
| CN107229535A (en) | Many copy storage methods of data block, storage device, method for reading data | |
| CN105159790A (en) | Data rescue method and file server | |
| CN102073655B (en) | A kind of method and apparatus preserving data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |