CN109491593A - A kind of data storage management system and method - Google Patents
A kind of data storage management system and method Download PDFInfo
- Publication number
- CN109491593A CN109491593A CN201811139586.4A CN201811139586A CN109491593A CN 109491593 A CN109491593 A CN 109491593A CN 201811139586 A CN201811139586 A CN 201811139586A CN 109491593 A CN109491593 A CN 109491593A
- Authority
- CN
- China
- Prior art keywords
- data file
- data
- flash memory
- converting unit
- physical 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0602—Interfaces specially adapted for storage systems specifically adapted to achieve a particular effect
- G06F3/061—Improving I/O performance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0628—Interfaces specially adapted for storage systems making use of a particular technique
- G06F3/0629—Configuration or reconfiguration of storage systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
- G06F3/06—Digital input from, or digital output to, record carriers, e.g. RAID, emulated record carriers or networked record carriers
- G06F3/0601—Interfaces specially adapted for storage systems
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/0671—In-line storage system
- G06F3/0683—Plurality of storage devices
- G06F3/0688—Non-volatile semiconductor memory arrays
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Techniques For Improving Reliability Of Storages (AREA)
Abstract
The invention discloses a kind of data storage management system and methods, belong to hardware store technical field;System includes flash memory converting unit, in the mapping processing and RAID checking treatment of the logical address and physical address for executing stored data file in flash memory converting unit, is set in the driver of computer;Writing unit, according to Physical Page from the data write request of user to flash memory converting unit application solid state hard disk physical address, to complete write operation;Reading unit reads corresponding data file to the request of flash memory converting unit according to the data read request of user, to complete read operation.The beneficial effect of above-mentioned technical proposal is: carrying out RAID verification and other calculating operations to solid state hard disk using the computing capability of host, improve the operation efficiency and readwrite performance of solid state hard disk, the idle computing capability of host is utilized simultaneously, but also even closer to the management of solid state hard disk reliable.
Description
Technical field
The present invention relates to hardware store technical field more particularly to a kind of data storage management system and methods.
Background technique
Solid state hard disk (Solid State Drives, SSD) is as a kind of emerging storage medium, relative to traditional
For mechanical hard disk, faster, shockproof to resist forthright lower, power consumption is lower, and operating noise is smaller for read or write speed, operating temperature range compared with
Greatly, and shape it is more traditional mechanical hard disk it is more frivolous.Therefore, with the development of memory technology, more and more storages are solved
Scheme is realized using solid state hard disk.
Existing high-end solid state hard disk is substantially the solid state hard disk based on flash memory, usually by solid-state hard disk controller and sudden strain of a muscle
Deposit chip composition, the RAID storage array of multiple flash chip composition solid state hard disks.Data file can quilt after entering RAID array
Multiple small data blocks are divided into, is then evenly distributed in multiple flash chips in RAID array and is saved.In this way
One group of small data block be thus referred to as a storage band.If store band in strip data where any one or it is more
A flash chip damage, then can be restored by strip data and verification data, to ensure the data safety of user.
In the existing solid state hard disk technology based on flash memory, usually held using the hard disk controller inside solid state hard disk
Row RAID verifies operation (XOR operation), does so that there are following disadvantages:
1) operational capability of solid-state hard disk controller is relatively weak, limits the performance of RAID operation, to limit solid
The readwrite performance of state hard disk;
2) host and solid state hard disk are mutually isolated, so that the internal logic of solid state hard disk can not be with host application program
It combines closely;
3) the idle operational capability of host is not used, host performance is caused to waste.
Summary of the invention
According to the above-mentioned problems in the prior art, the technical side of a kind of data storage management system and method is now provided
Case, it is intended to RAID verification and other calculating operations be carried out to solid state hard disk using the computing capability of host, improve solid state hard disk
Operation efficiency and readwrite performance, while the idle computing capability of host is utilized, but also more to the management of solid state hard disk
It is close reliable.
Above-mentioned technical proposal specifically includes:
A kind of data storage management system is applied in computer, is provided with solid state hard disk in the computer;Wherein,
Include:
Flash memory converting unit connects the solid state hard disk, stored data text is executed in Yu Suoshu flash memory converting unit
The logical address of part and the mapping processing of physical address, and execute the RAID checking treatment of the data file, the flash memory
Converting unit is set in the driver of the computer;
Writing unit, connects the flash memory converting unit, and said write unit is according to the data write request of user to institute
The physical address of the Physical Page of solid state hard disk described in flash memory converting unit application is stated, to complete the data file to the solid-state
The write operation of hard disk;
Reading unit, connects the flash memory converting unit, and the reading unit is according to the data read request of user to institute
It states the request of flash memory converting unit and reads the corresponding data file, to complete to read the data text from the solid state hard disk
The read operation of part.
Preferably, data storage management system, wherein the flash memory converting unit specifically includes:
Memory module, for saving the mapping table mapped between the logical address and the physical address;
Distribution module is written, connects the memory module, for being operated according to the said write of said write unit, to this
The secondary physical address for needing the data file being written to distribute corresponding multiple Physical Page in the solid state hard disk, said write
Distribution module writes the mapping relations between the logical address and physical address of the data file after distributing physical address
Enter in the mapping table;
Correction verification module is written, connects said write distribution module, for carrying out the RAID verification to the data file
Processing, the verification data of the data file are calculated;
Writing module connects said write correction verification module, for that will pass through the data text of the RAID checking treatment
Part and the corresponding verification data are stored in multiple Physical Page according to assigned physical address.
Preferably, data storage management system, wherein in the flash memory converting unit further include:
First judgment module is separately connected said write distribution module and said write correction verification module, for according to
The said write of writing unit operates, and judges that said write distribution module distributes whether the physical address succeeds, and output pair
The first judging result answered;
First cue module connects the first judgment module, for indicating described in distribution in first judging result
When physical address fails, to the first prompt information of user feedback write operation exception;
Said write correction verification module indicates to carry out institute again when distributing physical address success in first judging result
State the RAID checking treatment of data file.
Preferably, data storage management system, wherein multiple storage bands are set in Yu Suoshu solid state hard disk, each
It include multiple Physical Page in the storage band;
In the flash memory converting unit further include:
Second judgment module connects the write module, during the data file is written in the write module,
Whether the Physical Page of the current storage band of second judgment module judgement has been fully written, and exports corresponding the
Two judging results;
When second judging result indicates whether the Physical Page for storing band has been fully written, said write
The verification data are written in the solid state hard disk module.
Preferably, data storage management system, wherein the flash memory converting unit specifically includes:
Memory module, for saving the mapping table mapped between the logical address and the physical address;
Third judgment module connects the memory module, for the read operation according to the reading unit, in institute
The physical address for searching whether the data file read there are this needs in mapping table is stated, and exports corresponding third
Judging result;
Read module is separately connected the memory module and the third judgment module, for judging to tie in the third
When fruit indicates the physical address of the data file read in the mapping table there are this needs, in Yu Suoshu mapping table
The physical address for needing the Physical Page of the data file read is obtained, and reads the data text from the physical address
Part;
Second cue module connects the third judgment module, for indicating the mapping in the third judging result
To user feedback clear data when the physical address of the data file read in table there is no this needs, to prompt user
Data file reads failure.
Preferably, data storage management system, wherein multiple storage bands are set in Yu Suoshu solid state hard disk, each
It include multiple Physical Page in the storage band;
When executing said write operation to the data file, the flash memory converting unit is by the data file piecemeal
It is stored in multiple Physical Page, and saves the verification data simultaneously;
In the flash memory converting unit further include:
4th judgment module connects the read module, is for judging that the read module reads the data file
No success, and export corresponding 4th judging result;
Recovery module connects the 4th judgment module, for indicating the data file in the 4th judging result
When reading failure, by the data and the verification data saved in storage band corresponding to the data file,
It calculates and restores the data file;
The data file through over recovery is fed back to user by the flash memory converting unit.
A kind of data storage and management method is applied in computer, is provided with solid state hard disk in the computer;Wherein,
Writing process and reading process including data file;
Said write process specifically includes:
Step A1, using a writing unit obtain user data write request, and according to the data write request to
The physical address of the Physical Page of solid state hard disk described in the flash memory converting unit application, to carry out write operation to data file;
Step A2, use one are set to the flash memory converting unit inside the driver of the computer for the data text
The logical address of part is mapped to the physical address of multiple Physical Page, and executes the RAID checking treatment of the data file, with
The data file piecemeal is stored in the solid state hard disk, with backed off after random;
The mapping table of the logical address and physical address of the data file is set in the flash memory converting unit;
The reading process specifically includes:
Step B1, using a reading unit obtain user data read request, and according to the data read request to
The corresponding data file is read in the flash memory converting unit request, to be read to the data file;
Step B2 obtains the physical address of the data file using the flash memory converting unit in the mapping table,
To read the data file from the solid state hard disk, with backed off after random.
Preferably, data storage and management method, wherein the step A2 is specifically included:
Step A21, the flash memory converting unit operate according to the said write of said write unit, need to be written to this
The data file distribute the physical address of corresponding multiple Physical Page in the solid state hard disk, after distributing physical address
The mapping is written in mapping relations between the logical address and physical address of the data file by the flash memory converting unit
In table;
Step A22, the flash memory converting unit carries out the RAID checking treatment to the data file, to be calculated
The verification data of the data file;
Step A23, the flash memory converting unit carries out the RAID checking treatment to the data file, to be calculated
The verification data of the data file;
Step A24, the flash memory converting unit is by the data file and correspondence Jing Guo the RAID checking treatment
The verification data be stored in multiple Physical Page according to assigned physical address, with backed off after random.
Preferably, data storage and management method, wherein before executing the step A22, following steps are first carried out
It is rapid:
Step A211, the flash memory converting unit judge whether the distribution physical address succeeds:
If it is not, then to the first prompt information of user feedback write operation exception;
If so, turning to the step A22.
Preferably, data storage and management method, wherein multiple storage bands are set in Yu Suoshu solid state hard disk, each
It include multiple Physical Page in the storage band;
In the step A24, Yu Suoshu flash memory converting unit saves the data file and the corresponding check number
Following step is executed during
Whether the Physical Page of the current storage band of the flash memory converting unit judgement has been fully written, and is working as
The verification data are written in the solid state hard disk when the preceding storage band has write full, to complete to current institute
State the write operation of storage band.
Preferably, data storage and management method, wherein the step B2 is specifically included:
Step B21, the flash memory converting unit is according to the read operation of the reading unit, in Yu Suoshu mapping table
Search whether the physical address of the data file read there are this needs:
If it does not exist, then step B23 is turned to;
Step B22, the flash memory converting unit obtain the object for needing the data file read in the mapping table
The physical address of page is managed, and reads from the physical address data file and to user feedback, with backed off after random;
Step B23, the flash memory converting unit are lost to user feedback clear data with prompting subscriber data file to read
It loses, with backed off after random.
Preferably, data storage and management method, wherein the step B22 is specifically included:
Step B221, the flash memory converting unit judge that the read module reads whether the data file succeeds:
If reading successfully, the data file read is fed back to user by the flash memory converting unit, with retrogressing
Out;
Step B222, the flash memory converting unit is by saving in storage band corresponding to the data file
Data and the verification data, calculate and restore the data file;
Step B223, the data file being resumed is fed back to user by the flash memory converting unit, with backed off after random.
The beneficial effect of above-mentioned technical proposal is:
1) a kind of data storage management system is provided, the school RAID can be carried out to solid state hard disk using the computing capability of host
It tests with other calculating operations, improves the operation efficiency and readwrite performance of solid state hard disk, while the idle calculating of host is utilized
Ability, but also even closer to the management of solid state hard disk reliable;
2) a kind of data storage and management method is provided, above-mentioned data storage management system can be supported to operate normally.
Detailed description of the invention
Fig. 1 is a kind of general structure schematic diagram of data storage management system in preferred embodiment of the invention;
Fig. 2 is in preferred embodiment of the invention, and on the basis of Fig. 1, the specific structure of flash memory converting unit is illustrated
Figure;
Fig. 3 is in preferred embodiment of the invention, in a kind of data storage and management method, the overall procedure of writing process
Schematic diagram;
Fig. 4 is in preferred embodiment of the invention, in a kind of data storage and management method, the overall procedure of reading process
Schematic diagram;
Fig. 5 is in preferred embodiment of the invention, on the basis of Fig. 3, the idiographic flow schematic diagram of step A2;
Fig. 6 is the step schematic diagram needed to be implemented before executing step A22 in preferred embodiment of the invention;
Fig. 7 is in preferred embodiment of the invention, on the basis of Fig. 4, the idiographic flow schematic diagram of step B2;
Fig. 8 is the idiographic flow schematic diagram of step B22 in preferred embodiment of the invention.
Specific embodiment
The present invention will be further explained below with reference to the attached drawings and specific examples, but not as the limitation of the invention.
According to the above-mentioned problems in the prior art, a kind of data storage management system is now provided, is applied to computer
It is interior, solid state hard disk is provided in computer.The data storage management system is specifically as shown in fig. 1, comprising:
Flash memory converting unit 1 connects solid state hard disk 4, patrols in executing stored data file in flash memory converting unit 1
The mapping processing of address and physical address is collected, and executes the RAID checking treatment of data file, flash memory converting unit 1 is set to
In the driver of computer 5, correspondingly, writing unit 2 described below and reading unit 3 are all set in the computer 5
In;
Writing unit 2, connects flash memory converting unit 1, and writing unit 2 is converted according to the data write request of user to flash memory
Unit 1 applies for the physical address of the Physical Page of solid state hard disk, to complete write operation of the data file to solid state hard disk;
Reading unit 3, connects flash memory converting unit 1, and reading unit 3 is converted according to the data read request of user to flash memory
Unit 1 requests to read corresponding data file, to complete to read the read operation of data file from solid state hard disk.
Specifically, in the present embodiment, the configuration of computer (hereinafter referred to host) connection solid state hard disk can form one
The user terminal of a personal use can also form a server, or even form storage server cluster.
In the present embodiment, a flash memory converting unit 1 is set in the driver of above-mentioned host, which can be
The driver of the application program of solid state hard disk is matched in host, or the included driver of host, the purpose is to
Utilize the operational capability in host standby time.Specifically, above-mentioned flash memory converting unit 1 mainly has the function of following two: 1)
Processing forms the mapping relations between the logical address and physical address of the data file to be stored, establishes and saves mapping table;
2) RAID operation is carried out to the data file in the RAID storage array of deposit solid state hard disk, to obtain the check number of data file
According to substitute the RAID processor active task for needing to be executed by the hard disk controller in solid state hard disk originally.
Further, above-mentioned flash memory converting unit 1 can actually be realized relevant function by the processor of host, specifically
For, the processor of host generates one or more relevant processing threads in the driver of host, and using processing line
Journey realizes the function of above-mentioned flash memory converting unit 1.
In the present embodiment, a writing unit 2 is also set up in above-mentioned host, the data write-in for obtaining user's input is asked
It asks, and physical address is applied for flash memory converting unit 1 according to data write request, with the write operation of complete paired data file.
The function of above-mentioned writing unit 2 can be executed by the processor of host in the relevant application program of data storage management in host
Data write-in functions realize.The data write request of user can be manually entered the acquisition (process being locally stored by user
In), it can also be obtained (during cloud storage) by remote data communication, details are not described herein.
In the present embodiment, a reading unit 3 is also set up in above-mentioned host, the reading data for obtaining user's input is asked
It asks, and corresponding physical address is retrieved into flash memory converting unit 2 according to data read request, with the reading of complete paired data file
Extract operation.Similar with above-mentioned writing unit 2, which equally can execute data by the processor in host and deposit
Storage manages the reading data function in relevant application program to realize.The acquisition of above-mentioned data read request is also written with data
Request similar, details are not described herein.
To sum up, in technical solution of the present invention, in the driver of host be arranged flash memory converting unit 1, i.e., will originally by
The RAID operation that the hard disk controller of solid state hard disk executes, which is put into the driver of host, to be executed, and is taken full advantage of not busy in host
The operational capability set, and hard disk controller is released, to improve the readwrite performance of solid state hard disk, and host can be more
It is closely combined with solid state hard disk, plays the role of better hard disk management.
In preferred embodiment of the invention, as shown in Figure 2, above-mentioned flash memory converting unit 1 is specifically included:
Memory module 11, for saving the mapping table mapped between logical address and physical address;
Distribution module 12 is written, connects memory module 11, for the write operation according to writing unit 2, to this needs
The physical address of corresponding multiple Physical Page, write-in distribution module 12 are distributing in the data file distribution solid state hard disk of write-in
It, will be in the mapping relations write-in mapping table between the logical address and physical address of data file after physical address;
Correction verification module 13, connection write-in distribution module 12, for carrying out RAID checking treatment to data file, in terms of is written
Calculation obtains the verification data of data file;
Writing module 14, connection write-in correction verification module 13, for the data file of RAID checking treatment and right will to be passed through
The verification data answered are stored in multiple Physical Page according to assigned physical address.
Specifically, in the present embodiment, after writing unit 2, which writes data into request, to be sent into flash memory converting unit 1, flash memory
Converting unit 1 first according to the data write request, distribute corresponding multiple in solid state hard disk by the data file to need to save
Physical Page, the data file is divided into multiple small data blocks to treat at this time.Subsequent flash memory converting unit 1 is by the data file
Logical address (logical address of each small data block of division) and the physical address of assigned multiple Physical Page are mapped,
And the mapping relations between the logical address of the data file and the physical address of assigned multiple Physical Page are saved supreme
It states in the mapping table in memory module 11.Then, the RAID that flash memory converting unit 1 executes data file, which is verified, to be calculated, with processing
The verification data of the data file are obtained, and according to the physical address of assigned multiple Physical Page, data to be saved will be needed
File block is saved, while above-mentioned verification data being saved in one or more flash chips of solid state hard disk.
In preferred embodiment of the invention, still as shown in Figure 2, in flash memory converting unit 1 further include:
First judgment module 15 is separately connected write-in distribution module 12 and write-in correction verification module 13, for single according to write-in
The write operation of member 2 judges to be written whether 12 allocated physical address of distribution module succeeds, and exports corresponding first judgement knot
Fruit;
First cue module 16 connects first judgment module 15, for indicating allocated physical address in the first judging result
When failure, to the first prompt information of user feedback write operation exception;
Write-in correction verification module 13 carries out data file when the first judging result indicates allocated physical address success again
RAID checking treatment.
Specifically, in the present embodiment, after sending data write request to flash memory converting unit 3, flash memory turns writing unit 2
The case where 3 allocated physical address of unit fails there may be distribution is changed, the above-mentioned needs of first judgment module 15 first determine whether to distribute
Whether physical address succeeds, namely judges that writing unit 2 applies for whether physical address succeeds:
If apply physical address failure, then it represents that this data file storage failure, at this time the first cue module 16 to
The first prompt information of a write operation exception is fed back at family, and the process of this data file write-in terminates.
If applying for physical address success, enter the writing process of normal data file, write-in correction verification module 13 starts
The operation of RAID storage array is carried out, details are not described herein.
In preferred embodiment of the invention, multiple storage bands are set in solid state hard disk, are wrapped in each storage band
Include multiple Physical Page;
Then still as shown in Figure 2, in flash memory converting unit 1 further include:
Second judgment module 17 connects writing module 14, and during data file is written in writing module 14, second sentences
Disconnected module 17 judges whether the Physical Page of current storage band has been fully written, and exports corresponding second judging result;
When the second judging result indicates whether the Physical Page for storing band has been fully written, writing module 14 will verify data
It is written in solid state hard disk.
The concept of above-mentioned storage band with it is identical in the prior art, details are not described herein.
Specifically, in the present embodiment, during above-mentioned data file is stored in solid state hard disk, above-mentioned second judges mould
Block 17 persistently monitors whether the Physical Page that the storage item for being currently assigned to save the data file takes has been fully written.When current
Assigned storage band has been write completely, then data will be verified by, which demonstrating the need for, writes in flash chip, at this time flash memory converting unit 1
One or more flash chips verification data corresponding to strip data in the storage band being written in solid state hard disk
In.
In preferred embodiment of the invention, still as shown in Figure 2, flash memory converting unit 1 is specifically included:
Memory module 11, for saving the mapping table mapped between logical address and physical address;
Third judgment module 18 connects memory module 11, for the read operation according to reading unit 2, in mapping table
It searches whether the physical address of the data file read there are this needs, and exports corresponding third judging result;
Read module 19 is separately connected memory module 11 and third judgment module 18, for indicating in third judging result
When the physical address of the data file read in mapping table there are this needs, the data for needing to read are obtained in mapping table
The physical address of the Physical Page of file, and data file is read from physical address;
Second cue module 20 connects third judgment module 18, for indicating not deposit in mapping table in third judging result
In the physical address for the data file that this needs is read to user feedback clear data, to prompt subscriber data file to read
Failure.
Specifically, in the present embodiment, reading unit 3 sends data read request in flash memory converting unit 1, is read
The first step of extract operation is whether the data file for judging to need to read according to data read request has correlation in the mapping table
Physical address, in other words, it is necessary first to judge to need whether the data file that reads is already saved in solid state hard disk:
If being searched in the mapping table less than relevant physical address, then it represents that the data file is not stored in solid state hard disk
In or mapping table in there are certain mistakes, the second cue module 20 is to user feedback clear data at this time, to prompt user
Reading data failure.
If can search to obtain relevant physical address in the mapping table, read module 19 is directly from the physical address
It obtains data file and feeds back to user.
In preferred embodiment of the invention, multiple storage bands are set in solid state hard disk, are wrapped in each storage band
Include multiple Physical Page, the concept of above-mentioned storage band with it is above identical, details are not described herein.
When executing write operation to data file, data file piecemeal is stored in multiple Physical Page by flash memory converting unit
In, and verification data are saved simultaneously;
Then still as shown in Figure 2, in above-mentioned flash memory converting unit 1 further include:
4th judgment module 21 connects read module 19, for judging that read module 19 reads whether data file succeeds,
And export corresponding 4th judging result;
Recovery module 22 connects the 4th judgment module 21, for indicating that data file reads failure in the 4th judging result
When, by the data and verification data saved in storage band corresponding to data file, calculates and restore data file;
Data file through over recovery is fed back to user by above-mentioned flash memory converting unit 1.
The process that above-mentioned data are restored with it is similar in the prior art, i.e., by the strip data in storage band and relatively
The verification data answered restore the data in the flash chip of failure, to feed back to the complete data file of user.With existing skill
Unlike in art, the calculating and processing that above-mentioned data are restored execute in flash memory converting unit 1, i.e., in the driver of host
Middle progress, it makes use of the operational capabilities that host leaves unused, and the processor active task of the hard disk controller of solid state hard disk are released, to mention
The readwrite performance of solid state hard disk is risen.
In preferred embodiment of the invention, based on the above data storage management system, a kind of number is now provided
According to memory management method, it is applied in computer, solid state hard disk is provided in computer.
The above method includes the writing process and reading process of data file;
As shown in Figure 3, above-mentioned writing process specifically includes:
Step A1 obtains the data write request of user using a writing unit, and according to data write request to flash memory
The physical address of the Physical Page of converting unit application solid state hard disk, to carry out write operation to data file;
Step A2, use one are set to the flash memory converting unit inside the driver of computer for the logic of data file
Address of cache and executes the RAID checking treatment of data file at the physical address of multiple Physical Page, by data file point
Block is stored in solid state hard disk, with backed off after random;
The mapping table of the logical address and physical address of data file is set in flash memory converting unit;
As shown in Figure 4, above-mentioned reading process specifically includes:
Step B1 obtains the data read request of user using a reading unit, and according to data read request to flash memory
Corresponding data file is read in converting unit request, to be read to data file;
Step B2 obtains the physical address of data file using flash memory converting unit in the mapping table, with from solid state hard disk
Middle reading data file, with backed off after random.
In preferred embodiment of the invention, as shown in Figure 5, above-mentioned steps A2 is specifically included:
Step A21, write operation of the flash memory converting unit according to writing unit, the data file point that this needs is written
Physical address with multiple Physical Page corresponding in solid state hard disk, flash memory converting unit is literary by data after distributing physical address
In mapping relations write-in mapping table between the logical address and physical address of part;
Step A22, flash memory converting unit carries out RAID checking treatment to data file, the school of data file is calculated
Test data;
Step A23, flash memory converting unit carries out RAID checking treatment to data file, the school of data file is calculated
Test data;
Step A24, flash memory converting unit by Jing Guo RAID checking treatment data file and corresponding verification data according to
It is stored in multiple Physical Page according to assigned physical address, with backed off after random.
In preferred embodiment of the invention, before executing step A22, following steps as shown in Figure 6 are first carried out
It is rapid:
Step A211, flash memory converting unit judge whether allocated physical address succeeds:
If it is not, then to the first prompt information of user feedback write operation exception;
If so, turning to step A22.
In preferred embodiment of the invention, multiple storage bands are set in solid state hard disk, are wrapped in each storage band
Include multiple Physical Page;
In step A24, under being executed during flash memory converting unit saves data file and corresponding verification data
State step
Whether the Physical Page of the current storage band of flash memory converting unit judgement has been fully written, and in current storage band
Verification data are written in solid state hard disk when having write full, to complete the write operation to current storage band.
In preferred embodiment of the invention, as shown in Figure 7, above-mentioned steps B2 is specifically included:
Step B21, flash memory converting unit search whether that there are this according to the read operation of reading unit in mapping table
The physical address of the secondary data file for needing to read:
If it does not exist, then step B23 is turned to;
Step B22, flash memory converting unit obtain the Physical Page for the data file that needs are read physically in mapping table
Location, and data file and to user feedback is read from physical address, with backed off after random;
Step B23, flash memory converting unit is to user feedback clear data, to prompt subscriber data file to read failure, with
Backed off after random.
In preferred embodiment of the invention, as shown in Figure 8, above-mentioned steps B22 is specifically included:
Step B221, flash memory converting unit judge that read module reads whether data file succeeds:
If reading successfully, the data file read is fed back to user by flash memory converting unit, with backed off after random;
Step B222, flash memory converting unit is by storing the data and verification that save in band corresponding to data file
Data calculate and restore data file;
Step B223, the data file being resumed is fed back to user by flash memory converting unit, with backed off after random.
The above is only preferred embodiments of the present invention, are not intended to limit the implementation manners and the protection scope of the present invention, right
For those skilled in the art, it should can appreciate that and all replace with being equal made by description of the invention and diagramatic content
It changes and obviously changes obtained scheme, should all be included within the scope of the present invention.
Claims (10)
1. a kind of data storage management system is applied in computer, is provided with solid state hard disk in the computer;Its feature exists
In, comprising:
Flash memory converting unit connects the solid state hard disk, executes stored data file in Yu Suoshu flash memory converting unit
The mapping of logical address and physical address is handled, and executes the RAID checking treatment of the data file, the flash memory conversion
Unit is set in the driver of the computer;
Writing unit, connects the flash memory converting unit, and said write unit is according to the data write request of user to the sudden strain of a muscle
The physical address of the Physical Page of solid state hard disk described in converting unit application is deposited, to complete the data file to the solid state hard disk
Write operation;
Reading unit, connects the flash memory converting unit, and the reading unit is according to the data read request of user to the sudden strain of a muscle
It deposits converting unit request and reads the corresponding data file, to complete to read the data file from the solid state hard disk
Read operation.
2. data storage management system as described in claim 1, which is characterized in that the flash memory converting unit specifically includes:
Memory module, for saving the mapping table mapped between the logical address and the physical address;
Distribution module is written, connects the memory module, for being operated according to the said write of said write unit, this is needed
The data file to be written distributes the physical address of corresponding multiple Physical Page in the solid state hard disk, said write distribution
Institute is written after distributing physical address, by the mapping relations between the logical address and physical address of the data file in module
It states in mapping table;
Correction verification module is written, connects said write distribution module, for carrying out the RAID checking treatment to the data file,
The verification data of the data file are calculated;
Writing module, connect said write correction verification module, for will pass through the data file of the RAID checking treatment with
And the corresponding verification data are stored in multiple Physical Page according to assigned physical address.
3. data storage management system as claimed in claim 2, which is characterized in that in the flash memory converting unit further include:
First judgment module is separately connected said write distribution module and said write correction verification module, for according to said write
The said write of unit operates, and judges that said write distribution module distributes whether the physical address succeeds, and export corresponding
First judging result;
First cue module connects the first judgment module, distributes the physics for indicating in first judging result
When address fails, to the first prompt information of user feedback write operation exception;
Said write correction verification module indicates to carry out the number again when distributing physical address success in first judging result
According to the RAID checking treatment of file.
4. data storage management system as claimed in claim 2, which is characterized in that multiple deposit is arranged in Yu Suoshu solid state hard disk
Band is stored up, includes multiple Physical Page in each storage band;
In the flash memory converting unit further include:
Second judgment module connects the write module, described during the data file is written in the write module
Whether the Physical Page of the current storage band of the second judgment module judgement has been fully written, and exports corresponding second and sentence
Disconnected result;
When second judging result indicates whether the Physical Page for storing band has been fully written, the write module
The verification data are written in the solid state hard disk.
5. data storage management system as described in claim 1, which is characterized in that the flash memory converting unit specifically includes:
Memory module, for saving the mapping table mapped between the logical address and the physical address;
Third judgment module connects the memory module, and for the read operation according to the reading unit, Yu Suoshu reflects
The physical address of the data file read there are this needs is searched whether in firing table, and exports corresponding third judgement
As a result;
Read module is separately connected the memory module and the third judgment module, in the third judging result table
When showing the physical address of the data file read in the mapping table there are this needs, obtained in Yu Suoshu mapping table
The physical address of the Physical Page for the data file for needing to read, and the data file is read from the physical address;
Second cue module connects the third judgment module, for indicating in the mapping table in the third judging result
There is no this needs read the data file physical address when to user feedback clear data, to prompt user data
File reads failure.
6. data storage management system as claimed in claim 5, which is characterized in that multiple deposit is arranged in Yu Suoshu solid state hard disk
Band is stored up, includes multiple Physical Page in each storage band;
When executing said write operation to the data file, the flash memory converting unit saves the data file piecemeal
In multiple Physical Page, and the verification data are saved simultaneously;
In the flash memory converting unit further include:
4th judgment module connects the read module, for judge the read module read the data file whether at
Function, and export corresponding 4th judging result;
Recovery module connects the 4th judgment module, for indicating that the data file is read in the 4th judging result
When failure, by the data and the verification data saved in storage band corresponding to the data file, calculate
And restore the data file;
The data file through over recovery is fed back to user by the flash memory converting unit.
7. a kind of data storage and management method is applied in computer, is provided with solid state hard disk in the computer;Its feature exists
In writing process and reading process including data file;
Said write process specifically includes:
Step A1 obtains the data write request of user using a writing unit, and according to the data write request to described
The physical address of the Physical Page of solid state hard disk described in flash memory converting unit application, to carry out write operation to data file;
Step A2, use one are set to the flash memory converting unit inside the driver of the computer for the data file
Logical address is mapped to the physical address of multiple Physical Page, and executes the RAID checking treatment of the data file, by institute
It states data file piecemeal to be stored in the solid state hard disk, with backed off after random;
The mapping table of the logical address and physical address of the data file is set in the flash memory converting unit;
The reading process specifically includes:
Step B1 obtains the data read request of user using a reading unit, and according to the data read request to described
The corresponding data file is read in the request of flash memory converting unit, to be read to the data file;
Step B2 obtains the physical address of the data file using the flash memory converting unit in the mapping table, with from
The data file is read in the solid state hard disk, with backed off after random.
8. data storage and management method as claimed in claim 7, which is characterized in that the step A2 is specifically included:
Step A21, the flash memory converting unit are operated according to the said write of said write unit, the institute for need to be written to this
The physical address that data file distributes corresponding multiple Physical Page in the solid state hard disk is stated, it is described after distributing physical address
Mapping relations between the logical address and physical address of the data file are written in the mapping table flash memory converting unit;
Step A22, the flash memory converting unit carry out the RAID checking treatment to the data file, described to be calculated
The verification data of data file;
Step A23, the flash memory converting unit carry out the RAID checking treatment to the data file, described to be calculated
The verification data of data file;
Step A24, the flash memory converting unit by Jing Guo the RAID checking treatment the data file and corresponding institute
It states verification data to be stored in multiple Physical Page according to assigned physical address, with backed off after random;
Preferably, before executing the step A22, following step is first carried out:
Step A211, the flash memory converting unit judge whether the distribution physical address succeeds:
If it is not, then to the first prompt information of user feedback write operation exception;
If so, turning to the step A22;
Preferably, the step B2 is specifically included:
Step B21, the flash memory converting unit are searched in Yu Suoshu mapping table according to the read operation of the reading unit
With the presence or absence of the physical address for the data file for thering is this needs to read:
If it does not exist, then step B23 is turned to;
Step B22, the flash memory converting unit obtain the Physical Page for needing the data file read in the mapping table
Physical address, and the data file is read from the physical address and to user feedback, with backed off after random;
Step B23, the flash memory converting unit is to user feedback clear data, to prompt subscriber data file to read failure, with
Backed off after random.
9. data storage and management method as claimed in claim 8, which is characterized in that multiple deposit is arranged in Yu Suoshu solid state hard disk
Band is stored up, includes multiple Physical Page in each storage band;
In the step A24, Yu Suoshu flash memory converting unit saves the data file and the corresponding verification data
Following step is executed in the process;
Whether the Physical Page of the current storage band of the flash memory converting unit judgement has been fully written, and current
The verification data are written in the solid state hard disk when storage band has write full, to complete to depositing described in current
Store up the write operation of band.
10. data storage and management method as claimed in claim 8, which is characterized in that the step B22 is specifically included:
Step B221, the flash memory converting unit judge that the read module reads whether the data file succeeds:
If reading successfully, the data file read is fed back to user by the flash memory converting unit, with backed off after random;
Step B222, the flash memory converting unit pass through the data that save in storage band corresponding to the data file
And the verification data, it calculates and restores the data file;
Step B223, the data file being resumed is fed back to user by the flash memory converting unit, with backed off after random.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811139586.4A CN109491593B (en) | 2018-09-28 | 2018-09-28 | Data storage management system and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811139586.4A CN109491593B (en) | 2018-09-28 | 2018-09-28 | Data storage management system and method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109491593A true CN109491593A (en) | 2019-03-19 |
CN109491593B CN109491593B (en) | 2020-11-17 |
Family
ID=65690057
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811139586.4A Active CN109491593B (en) | 2018-09-28 | 2018-09-28 | Data storage management system and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109491593B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111538460A (en) * | 2020-04-20 | 2020-08-14 | 江苏芯盛智能科技有限公司 | RAID function implementation method and related device |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101576853A (en) * | 2008-05-06 | 2009-11-11 | 群联电子股份有限公司 | Data access method and controller and memory system using same |
US20100146194A1 (en) * | 2008-12-05 | 2010-06-10 | Apacer Technology Inc. | Storage Device And Data Management Method |
CN102541760A (en) * | 2012-01-04 | 2012-07-04 | 记忆科技(深圳)有限公司 | Computer system based on solid-state hard disk |
CN103049222A (en) * | 2012-12-28 | 2013-04-17 | 中国船舶重工集团公司第七0九研究所 | RAID5 (redundant array of independent disk 5) write IO optimization processing method |
CN103999060A (en) * | 2011-12-23 | 2014-08-20 | 国际商业机器公司 | Solid-state storage management |
CN107870869A (en) * | 2016-09-28 | 2018-04-03 | 三星电子株式会社 | Include the computing system of the storage device by host computer control |
CN108073470A (en) * | 2016-11-07 | 2018-05-25 | 三星电子株式会社 | Perform the storage system and its control method of the error correction of address mapping table |
CN108509353A (en) * | 2018-03-14 | 2018-09-07 | 清华大学 | Object storage construction method based on naked flash memory and device |
-
2018
- 2018-09-28 CN CN201811139586.4A patent/CN109491593B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101576853A (en) * | 2008-05-06 | 2009-11-11 | 群联电子股份有限公司 | Data access method and controller and memory system using same |
US20100146194A1 (en) * | 2008-12-05 | 2010-06-10 | Apacer Technology Inc. | Storage Device And Data Management Method |
CN103999060A (en) * | 2011-12-23 | 2014-08-20 | 国际商业机器公司 | Solid-state storage management |
CN102541760A (en) * | 2012-01-04 | 2012-07-04 | 记忆科技(深圳)有限公司 | Computer system based on solid-state hard disk |
CN103049222A (en) * | 2012-12-28 | 2013-04-17 | 中国船舶重工集团公司第七0九研究所 | RAID5 (redundant array of independent disk 5) write IO optimization processing method |
CN107870869A (en) * | 2016-09-28 | 2018-04-03 | 三星电子株式会社 | Include the computing system of the storage device by host computer control |
CN108073470A (en) * | 2016-11-07 | 2018-05-25 | 三星电子株式会社 | Perform the storage system and its control method of the error correction of address mapping table |
CN108509353A (en) * | 2018-03-14 | 2018-09-07 | 清华大学 | Object storage construction method based on naked flash memory and device |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111538460A (en) * | 2020-04-20 | 2020-08-14 | 江苏芯盛智能科技有限公司 | RAID function implementation method and related device |
Also Published As
Publication number | Publication date |
---|---|
CN109491593B (en) | 2020-11-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10778751B2 (en) | Method of micro-service transformation for power trading functions | |
US9128872B2 (en) | Techniques for providing data redundancy after reducing memory writes | |
US8103825B2 (en) | System and method for providing performance-enhanced rebuild of a solid-state drive (SSD) in a solid-state drive hard disk drive (SSD HDD) redundant array of inexpensive disks 1 (RAID 1) pair | |
CN101727299B (en) | RAID5-orientated optimal design method for writing operation in continuous data storage | |
US7904647B2 (en) | System for optimizing the performance and reliability of a storage controller cache offload circuit | |
CN107357680A (en) | Data processing equipment and the method for data processing | |
US20130282672A1 (en) | Storage apparatus and storage control method | |
CN104407933B (en) | A kind of backup method and device of data | |
CN109358809B (en) | RAID data storage system and method | |
US9785514B1 (en) | Techniques for file system recovery | |
CN101436149B (en) | Method for rebuilding data of magnetic disk array | |
CN106227731B (en) | A kind of data processing method, device and distributed file system | |
CN106066890B (en) | Distributed high-performance database all-in-one machine system | |
CN103516736A (en) | Data recovery method of distributed cache system and a data recovery device of distributed cache system | |
CN107729536B (en) | Data storage method and device | |
CN103049508B (en) | A kind of data processing method and device | |
CN107273046B (en) | A kind of data processing method and system based on Solid-state disc array | |
CN103049220A (en) | Storage control method, storage control device and solid-state storage system | |
CN102177496A (en) | System and method for transferring data between different RAID data storage types for current data and replay data | |
WO2021139166A1 (en) | Error page identification method based on three-dimensional flash storage structure | |
CN104050014B (en) | Efficient storage management method based on virtual platform | |
CN108874312B (en) | Data storage method and storage device | |
CN103699340B (en) | A kind of request processing method and equipment | |
CN106919339B (en) | Hard disk array and method for processing operation request by hard disk array | |
US20210318739A1 (en) | Systems and methods for managing reduced power failure energy requirements on a solid state drive |
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 | ||
PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A data storage management system and method Effective date of registration: 20230518 Granted publication date: 20201117 Pledgee: Jiangsu Bank Co.,Ltd. Shanghai Huinan Branch Pledgor: FANGYI INFORMATION TECHNOLOGY (SHANGHAI) CO.,LTD. Registration number: Y2023310000195 |
|
PE01 | Entry into force of the registration of the contract for pledge of patent right |