CN102053919A - Data dependency identification method and memory data processing method - Google Patents
Data dependency identification method and memory data processing method Download PDFInfo
- Publication number
- CN102053919A CN102053919A CN2009102076682A CN200910207668A CN102053919A CN 102053919 A CN102053919 A CN 102053919A CN 2009102076682 A CN2009102076682 A CN 2009102076682A CN 200910207668 A CN200910207668 A CN 200910207668A CN 102053919 A CN102053919 A CN 102053919A
- Authority
- CN
- China
- Prior art keywords
- memory
- data
- block
- disposal route
- pages
- 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.)
- Pending
Links
Images
Abstract
The invention discloses a data dependency identification method and a memory data processing method. Data with dependency are stored in the same storage block. In addition, the data with dependency are multiple data which are stored by a user so as to perform a specific function. In other words, the multiple data are used for performing the specific function.
Description
Technical field
The present invention relates to a kind of data dependence discrimination method and memory data disposal route thereof, relate in particular to a kind of correlation of data discrimination method and memory data disposal route thereof of consecutive access.
Background technology
The characteristic of flash memory can't directly override, and erasing times is limited.Can't directly override, be meant new data before storing, its corresponding flash memory storage block is if having data, could store new data after then need wiping these data earlier.In addition, the processing time of carrying out data erase in storer is longer than data read or time of writing, therefore, if can reduce the number of times of data erase then can promote the access usefulness of storer.
Flash memory also has the restriction of its erasing times, if therefore access data and often storer is wiped processing, then can reduce the serviceable life of storer.In known technology,, therefore, utilize the performance analysis data attribute that it is categorized as cold data (cold data) and dsc data (hot data) in order to reduce the moving of data, erasing times.So-called hot and cold data are meant the data of less access in a unit interval and the data of normal access respectively.Known technology by with cold data storage in same memory block, dsc data is stored in the mode of another memory block, makes hot and cold data can handle respectively promoting access efficiency, and reduces required erasing times.Yet, classified as hot and cold data according to data by the frequent degree of access and still have its shortcoming, therefore, proposing a kind of new memory data disposal route and discrimination method has its necessity in fact.
Summary of the invention
A purpose of the present invention is to propose a kind of memory data disposal route.This method will have the data storage of correlativity in the same memory region piece, therefore if need then needn't carry out access in a plurality of memory block during these data of access.Wherein, the data with correlativity are meant many data of carrying out access in regular turn according to ad hoc rules, or the user is for carrying out many data of specific function consecutive access.
Another object of the present invention is to propose a kind of method of differentiating correlativity between data.Wherein, the data with correlativity are meant the user for carrying out many data of specific function consecutive access, and therefore, these many data are applied to carrying out this specific function.
Propose a kind of memory data disposal route in an embodiment of the present invention, wherein this storer comprises a plurality of memory block, and each memory block has a plurality of memory pages, and this data processing method comprises:
(a) from a plurality of memory block, find out the data storage block of written data;
(b), a plurality of memory pages of consecutive access are wherein classified as at least one memory page group to a plurality of memory pages of at least one data storage block; And
(c) a plurality of memory pages with at least one memory page group are copied to the blank memory block that does not write data in a plurality of memory block.
A kind of data dependence discrimination method is proposed in an embodiment of the present invention, a plurality of data that have the data dependence of consecutive access in order to identification, each data in these a plurality of data correspond respectively to the logical block addresses in the storer, and this method comprises:
(a) logical block addresses of at least one data correspondence in a plurality of the data is carried out computing via a function, export one first numerical value;
(b) in a plurality of data after these at least one data another data at least of access, the logical block addresses that it is corresponding is carried out computing via this function, exports a second value;
(c) relative coefficient and the third value with first, second numerical value correspondence adds up; And
(d) when this relative coefficient greater than a critical value, judge that the corresponding respectively at least one data of first, second numerical value and another data at least have the data dependence of consecutive access.
Description of drawings
Fig. 1 shows the physical blocks address of known as memory device and the synoptic diagram of logical block addresses.
Fig. 2 shows the embodiment process flow diagram of memory data disposal route of the present invention.
Fig. 3 display-memory is used the synoptic diagram of notebook data disposal route.
Fig. 4 shows data dependence discrimination method process flow diagram of the present invention.
The synoptic diagram of Fig. 5 a and Fig. 5 b display application data dependence discrimination method of the present invention.
[main element symbol description]
Flash memory 310
Memory page group 320
Data data1, data2, data3, data4, data5, data7, data8,
data11、data26
Relative coefficient Locality (K
1, H
1), Locality (K
2, H
1), Locality (K
3,
H
1)、Locality(H
1，H
2)、Locality(K
2，H
2)、Locality(K
3，H
2)
Relative coefficient table Locality
Function Hash
Embodiment
Because known technology is to be hot and cold data with data qualification, makes many data to be stored in different memory block respectively and reduce the data access efficiency of storer with data dependence.The present invention proposes a kind of data dependence discrimination method, and the memory data disposal route.The embodiment of the invention can be judged the data of the correlativity with consecutive access, and can with data storage with correlativity in storer identical or close memory block to promote data access efficiency.
Fig. 1 shows the physical blocks address of a known as memory device and the synoptic diagram of logical block addresses.As shown in Fig. 1, include a plurality of memory block (Block) in the storer 110, comprise a plurality of memory pages (page) that are used to store data in each memory block.Each memory page all have a correspondence the physical blocks address (Physical BlockAddress, PBA), and each physical blocks address also corresponding to a logical block addresses (Logical Block Address, LBA).Continuous physical blocks address PBA may not be corresponding to continuous logical block addresses.External system (scheming not shown) is to the process of storer 110 access datas, be to send the control module (scheme not shown) of a logical block addresses earlier, inquire about this logical block addresses corresponding physical blocks address in storer 110 by control module then to storer 110.At last, data on this physical blocks address of control module access and send external system to.
As discussed previously, in order to improve the access efficiency of storer, so data are divided into hot and cold data according to the access times in the unit interval, and dsc data is stored in same memory block, and cold data storage is in another memory block in the known technology.Wherein, dsc data is the data of frequent access comparatively, and cold data then are the data of less access.Yet many dsc datas in the same memory block may not have the correlativity that the present invention censures each other.Therefore as user during for the dsc data carrying out specific function and need to be distinguished under many known technologies of access, may need among different memory block these dsc datas of access and reduced access efficiency.
In view of this, a kind of memory data disposal route has been proposed among the present invention.This method will have the data storage of correlativity of consecutive access in the same memory region piece, therefore needn't distinguish these data of access in a plurality of memory block.In embodiments of the present invention, has the data representation user of correlativity of consecutive access for carrying out many data of specific function access.In other words, these many data are applied to carrying out this specific function.For example the user is the object computer start, and the data that then have the correlativity of consecutive access comprise bios program code, disk boot-strap program code, os starting program code or the like.Therefore, bios program code, disk boot-strap program code, os starting program code promptly have the data of the correlativity of consecutive access.For example when the data folder and the file of user's desire inquiry memory storage, the data that then have the correlativity of consecutive access comprise data folder name data, subdata folder name data again, and the file name data or the like.In addition, consecutive access is the order of steps consecutive accesss of these data of expression according to the execution specific function, but not is defined in the consecutive access on the express time.
The data that two kinds of discrimination methods are applied to differentiate correlativity with consecutive access have been proposed in the embodiment of the invention.Wherein the method characteristic of utilizing storer often to write data continuously according to the logical block addresses order is carried out identification.And the opposing party's rule utilizes statistical to come whether to have between the Identification Data correlativity of consecutive access.
Fig. 2 shows an embodiment process flow diagram of memory data disposal route among the present invention.This method is the data processing method that is applied to a storer, and this storer is a flash memory in the present embodiment.This flash memory comprises a plurality of memory block, and each memory block all has a plurality of memory pages, and each memory page also has corresponding logical block addresses and physical blocks address.At first, in a plurality of memory block of flash memory, find out the data storage block (S110) of written data earlier.Expended for a long time for fear of the data storage block of finding out all written data, embodiment can set and find the data storage block of a given number (for example at least one) promptly to carry out next step.
Then, a plurality of memory pages that wherein are recognized as the energy consecutive access are classified as at least one memory page group (S120) to a plurality of memory pages at least one data storage block.The method of a plurality of memory pages of the energy of identification in the present embodiment consecutive access, be that the data of utilizing storer often will have correlativity write the characteristic of memory page according to the logical block addresses order, so the data identification that will be stored in the continuous logic block address is the correlativity with consecutive access.Therefore, this method at first writes down the logical block addresses of a plurality of memory pages of this at least one data storage block; The close a plurality of memory pages of logical block addresses are referred at least one memory page group in then will these a plurality of memory pages.Wherein, the numerical value difference of the close presentation logic block address of logical block addresses is in a particular range, and for example numerical value difference is no more than 3.Among another embodiment, the numerical value of the close presentation logic block address of logical block addresses is continuous, and for example the logical block addresses of a plurality of memory pages is respectively 1,2,3.
Be recognized as can consecutive access a plurality of memory pages be classified as at least one memory page group after, then just a plurality of memory pages of memory page group are copied to the blank memory block (S130) that does not write data in aforementioned a plurality of memory block.And can be apace by access for a plurality of memory pages that make same memory page group, these memory pages are written in the same blank memory block.And write the characteristic of data in order to meet flash memory according to the logical block addresses order, classified a plurality of memory pages to same memory page group are written in the same blank memory block according to the logical block addresses order.
In addition, optionally comprise the data (S140) of wiping in the aforementioned data memory block in the present embodiment, provide follow-up data storage to make this data storage block become blank memory block.Above-described memory data disposal route can be by the manual activation of a user, or this flash memory is promptly carried out this method every a special time.
Fig. 3 shows the synoptic diagram of a memory application notebook data disposal route.As shown in the figure, comprise many memory block block 0, block 1, block 2, block n and other memory block or the like in one flash memory 310, and each memory block comprises a plurality of memory page page 0, page 1, page 2, page 3, page 4.Shown in Fig. 3 first half, the data storage block of the memory block block 0 in the flash memory 310, block 1 written data, memory block block 2, block n then are the blank memory block that does not write data.When flash memory 310 is used the notebook data disposal route, then at first can find out the data storage block of a given number written data, promptly carry out next step after setting the data storage block (for example data storage block block 0) find a written data in this example according to step (S110).
Then according to step (S120), be recognized as among the data storage block block 0 can consecutive access a plurality of memory pages classify as at least one memory page group 320.One of feature of the memory page of energy consecutive access is that these memory pages have continuous logical block addresses order.Form this feature reason since storer usually according to logical block addresses order continuity write data.Therefore, the logical block addresses 7,2,3,26,1 of all memory page page 0, the page 1 of first record data memory block block 0, page 2, page3, page 4 in step (S120); Then a plurality of memory pages that logical block addresses is close among these memory pages page 0, page1, page 2, page 3, the page 4 are referred to memory page group 320.Especially, be will have a plurality of memory page page4, the page 1 of continuous logic block address 1,2,3, the information of page 2 (comprising logical block addresses, physical blocks address) is recorded in the memory page group 320 according to the logical block addresses order, the information with remaining memory page page 0, page 3 is arranged in thereafter again.Step (S120) mainly is to be recorded in same memory page group for the information of a plurality of memory pages that will have the continuous logic block address according to the logical block addresses order, therefore in another embodiment, the information of the memory page (for example memory page page 3, page 4) of discontinuous logical block addresses can be stored in another memory page group.
At last, step (S130) is copied to the memory page of a blank memory block according to the logical block addresses order with the data of each memory page according to a plurality of memory page information of record in the memory page group 320, and this blank memory block is the memory block that does not write data.Be to pick out the new data memory block of blank memory block block n in this example as the storage data, therefore shown in Fig. 3 Lower Half, step (S130) is according to each memory page logical block addresses 1,2,3,7,26 order of storage in the memory page group 320, duplicates its corresponding data data1, data2, data3, data7, data26 to memory page page 0, the page 1 of new data memory block block n, page 2, page 3, page 4.Step (S130) mainly is that data storage for a plurality of memory pages that will have the continuous logic block address is in same memory block.Therefore among another embodiment, the data of the memory page (for example memory page page 3, page 4) of discontinuous logical block addresses can be stored in another memory block.In addition, because the data among the data storage block block 0 are stored in data storage block block n in addition according to this method embodiment, therefore can become a new blank memory block to make it according to the data among step (S140) the obliterated data memory block block 0.
A kind of memory data disposal route proposed by the invention is more than disclosed, the data storage of correlativity that this method will have a consecutive access identical or close memory block in storer.The foregoing description is only put down in writing the processing procedure of a data memory block, known those skilled in the art when understanding if handle a plurality of data storage blocks simultaneously, then more can be with the data storage that wherein has a correlativity in identical or close memory block to promote access efficiency.In addition, the method that a kind of identification has the correlation data of consecutive access has above also been described.This discrimination method utilizes storer often to write the characteristic of data continuously according to the logical block addresses order, so the data identification that will be stored in the continuous logic block address is the correlativity with consecutive access.In addition, the embodiment of the invention has also proposed another kind of discrimination method, and it utilizes statistical to come whether to have between the Identification Data correlativity of consecutive access.
Fig. 4 shows the data dependence discrimination method process flow diagram that the embodiment of the invention proposes, and it utilizes statistical to come identification to have the data of the correlativity of consecutive access.Each data in a plurality of data correspond respectively to the logical block addresses in the storer.At first, at least one data desire in these a plurality of data begins to carry out access, and then that it is corresponding logical block addresses carries out exporting one first numerical value (S410) after the computing via a function, and wherein, each logical block addresses all is a numerical value.In one embodiment, can be according to the order of data access, the logical block addresses of several data of access is at first carried out computing via this function, then the indivedual first corresponding numerical value of output.In addition, this function can be a hash function, a remainder function, or other function.
Then, in these a plurality of data after these at least one data another data at least of access, the logical block addresses that it is corresponding is carried out computing via same function and is exported a second value (S420).In step (S420), this at least another data be to continue after these at least one data by the data of access, therefore between the two may the person of being to use for carrying out the data of specific function consecutive access, but still need remake further judgement.Next, a relative coefficient of this first, second numerical value correspondence and a third value are added up (S430).Wherein, whether another data are often by consecutive access at least to be used to these at least one data of the amount of commenting and this corresponding to the relative coefficient of first, second numerical value, and therefore this relative coefficient just can add up with third value (for example numerical value is 1) when the situation of preface access takes place to connect.
At last, this relative coefficient is compared with a critical value, when relative coefficient just can be judged these at least one data and this data dependence (S440) that has consecutive access at least between another data that this first, second numerical value is corresponding respectively during greater than critical value.In addition, this discrimination method can optionally comprise step (S450), in order to reduce the numerical value of relative coefficient, its objective is for the custom that reduces the previous access data of user to the current influence of doing judgement.And reduce the method for relative coefficient numerical value, can for example deduct the numerical value of this relative coefficient with the 4th numerical value, or with this relative coefficient divided by the 5th numerical value.And above-described step (S450) can be by the manual activation of a user, or promptly carries out this step every a special time.
The synoptic diagram of Fig. 5 a and Fig. 5 b display application data dependence discrimination method of the present invention.Shown in Fig. 5 a, in step (S410) when many data are carried out consecutive access, the data dependence that whether has consecutive access for these data of identification, then at first with at least one data of access the earliest, for example with the numerical value 21,25,4 of the logical block addresses of three data data1, data2 of access at first, data3, carry out corresponding three first numerical value K1=1, K2=2, the K3=4 of output after the computing via a function Hash according to its access order.
Then, according to the logical block addresses of step (S420) to another data at least of access after these three data data1, data2, the data3, for example the numerical value 7 of the logical block addresses of data data4 carries out exporting corresponding second value H1=3 after the computing via same function Hash.Next, relative coefficient Locality that these first numerical value are corresponding with this second value in the step (S430) (K1, H1), Locality (K2, H1), Locality (K3, H1) add up with a third value respectively, this third value is 1 in the present embodiment.For the ease of writing down a plurality of relative coefficients, can adopt a relative coefficient table Locality to write down the numerical value of each relative coefficient in the present embodiment.For example K1=1, H1=3, thus Locality (K1, numerical value H1) are recorded in promptly that the Building X scale value is 1 among the relative coefficient table Locality, and the Building Y scale value is in 3 the field.
Afterwards, step (S440) with each relative coefficient Locality (K1, H1), Locality (K2, H1), (K3's Locality H1) compares with a critical value.For example this critical value is made as 100, when the numerical value of relative coefficient can judge that just pairing two data of this relative coefficient have the data dependence of consecutive access greater than 100 the time.For instance, if (K1, numerical value H1) then represented the data data1 of the logical block addresses 21 of K1 correspondence greater than 100 o'clock to relative coefficient Locality, and the data data4 of the logical block addresses 7 corresponding with H1 has the data dependence of consecutive access.In addition, be that numerical value with K1=1 and H1=3 carries out computing in the hope of its corresponding logical block addresses 21,7 via an inverse function in step (S440), and this inverse function correspond to aforesaid function.
Known those skilled in the art be when can understanding when many data are carried out access, and abovementioned steps (S410) to step (S440) can repeat the data of using the access of assessment next record and the consecutive access correlativity between the past data.Shown in Fig. 5 b, when next data data5 desires to carry out access, then at first calculate first numerical value of its first three data according to step (S410).Wherein numerical value 25,4, the 7 interdependent order of getting of the logical block addresses of data data2, data3, data4 carry out exporting corresponding three first numerical value K2=2, K3=4, H1=3 after the computing via this function Hash.Also or the first numerical value H1 that the logical block addresses 7 of data data4 is exported after function Hash computing, replacing the first numerical value K1 that the logical block addresses 21 of data data1 exports after function carries out computing gets final product.According to step (S420) logical block addresses 33 of the data data5 of access after these three data data2, data3, the data4 is carried out exporting corresponding second value H2=1 after the computing via same function then.
Relative coefficient Locality that these first numerical value are corresponding with this second value in the step (S430) (H1, H2), Locality (K2, H2), (K3's Locality H2) adds up with a third value 1 respectively.At last, step (S440) with each relative coefficient Locality (H1, H2), Locality (K2, H2), (K3's Locality H2) compares with critical value.And just can judge the data dependence that has consecutive access between pairing two data of this relative coefficient during greater than critical value when the numerical value of relative coefficient.More than explanation only is the preferred embodiments of the present invention, is not in order to limit embodiments of the present invention.For example, can select for use other certain number destination data to ask for first numerical value, or other certain number destination data is asked for second value.In addition, the yet correlativity that can utilize alternate manner to come whether to have between the statistics consecutive access does not limit and only can use relative coefficient and set up the relative coefficient table and come correlativity between statistics.The above-described preferred embodiment of the present invention discloses wherein two kinds of discrimination methods, have the data of the correlativity of consecutive access in order to identification, but other discrimination method also belongs to claim of the present invention.Wherein, has the data representation user of correlativity of consecutive access for carrying out many data of specific function access.
The above is the preferred embodiments of the present invention only, is not in order to limit claim of the present invention; All other do not break away from the equivalence of being finished under the disclosed spirit and changes or modification, all should be included in the scope of protection of present invention.
Claims (13)
1. memory data disposal route, wherein this storer comprises a plurality of memory block, and each these a plurality of memory block has a plurality of memory pages, and this data processing method comprises:
(a) from these a plurality of memory block, find out the data storage block of written data;
(b), a plurality of memory pages of consecutive access are wherein classified as at least one memory page group to these a plurality of memory pages of at least one data storage block; And
(c) these a plurality of memory pages that will this at least one memory page group are copied to the blank memory block that does not write data in these a plurality of memory block.
2. memory data disposal route as claimed in claim 1 wherein comprises in this step (b):
(b1) write down the logical block addresses LBA of these a plurality of memory pages of this at least one data storage block; And
(b2) the close a plurality of memory pages of logical block addresses classify as this at least one memory page group in will these a plurality of memory pages.
3. memory data disposal route as claimed in claim 2, these a plurality of memory pages in this step (b2) have continuous logical block addresses.
4. memory data disposal route as claimed in claim 1, wherein this step (c) also comprises these a plurality of memory pages in this at least one memory page group, writes same blank memory block.
5. memory data disposal route as claimed in claim 1, wherein this step (c) also comprise according to should be at least one in the memory page group these a plurality of memory pages logical block addresses in proper order, write same should the blank memory block.
6. memory data disposal route as claimed in claim 1, wherein this memory data disposal route is by the manual activation of a user.
7. memory data disposal route as claimed in claim 1, wherein this storer is promptly carried out this memory data disposal route every a special time.
8. memory data disposal route as claimed in claim 1 also comprises:
(d) wipe data in this at least one data storage block.
9. memory data disposal route, wherein this storer comprises a plurality of memory block, and each these a plurality of memory block has a plurality of memory pages, and this data processing method comprises:
(a) from these a plurality of memory block, find out the data storage block of written data;
(b) logical block addresses of these a plurality of memory pages of at least one data storage block of record;
(c) the close a plurality of memory pages of logical block addresses classify as at least one memory page group in will these a plurality of memory pages;
(d) these a plurality of memory pages in will this at least one memory page group are copied to the blank memory block that does not write data in these a plurality of memory block; And
(e) wipe data in this at least one data storage block.
10. memory data disposal route as claimed in claim 9 wherein at least one memory page of this of this step (c) group, comprises these a plurality of memory pages with continuous logic block address.
11. memory data disposal route as claimed in claim 9, wherein this step (d) also comprise according to should be at least one in the memory page group these a plurality of memory pages logical block addresses in proper order, write same should the blank memory block.
12. memory data disposal route as claimed in claim 9, wherein this memory processes method is by the manual activation of a user.
13. memory data disposal route as claimed in claim 9, wherein this storer is promptly carried out this memory data disposal route every a special time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009102076682A CN102053919A (en) | 2009-10-29 | 2009-10-29 | Data dependency identification method and memory data processing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2009102076682A CN102053919A (en) | 2009-10-29 | 2009-10-29 | Data dependency identification method and memory data processing method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102053919A true CN102053919A (en) | 2011-05-11 |
Family
ID=43958264
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2009102076682A Pending CN102053919A (en) | 2009-10-29 | 2009-10-29 | Data dependency identification method and memory data processing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102053919A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107368258A (en) * | 2016-05-11 | 2017-11-21 | 慧荣科技股份有限公司 | Method for screening damaged data row of data storage medium |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101464836A (en) * | 2007-12-20 | 2009-06-24 | 宇瞻科技股份有限公司 | Data memory method for flash memory device |
-
2009
- 2009-10-29 CN CN2009102076682A patent/CN102053919A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101464836A (en) * | 2007-12-20 | 2009-06-24 | 宇瞻科技股份有限公司 | Data memory method for flash memory device |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107368258A (en) * | 2016-05-11 | 2017-11-21 | 慧荣科技股份有限公司 | Method for screening damaged data row of data storage medium |
| CN107368258B (en) * | 2016-05-11 | 2020-06-12 | 慧荣科技股份有限公司 | Method for screening damaged data row of data storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105843749B (en) | A kind of NAND Flash fault-tolerance approach based on FPGA | |
| CN103455432B (en) | It is used for managing the method for a memory and its relevant memory | |
| CN101419571A (en) | Method for storing configuration parameter in NOR FLASH based on Hash arithmetic | |
| TWI408551B (en) | Falsh memory storage system and flash memory controller and data processing method thereof | |
| CN102135942B (en) | Method for realizing wear-leveling in storage equipment as well as the storage equipment | |
| CN101446921B (en) | Dynamic storage method of Flash memory | |
| CN101640069B (en) | Average wear method and average wear system for flash memory | |
| CN1645516B (en) | Data recovery apparatus and method used for flash memory | |
| CN100382049C (en) | Data processing apparatus and method for flash memory | |
| WO2017161795A1 (en) | Method and system for processing chronological file | |
| CN103577574A (en) | High-reliability linear file system based on nand flash | |
| CN101515276A (en) | Method for write operation of file data, and recovery method and recovery system for file data | |
| CN101122924A (en) | Logic log generation method, database backup/ restoration method and system | |
| CN102737715A (en) | Data brown-out protection method for NOR flash memory | |
| CN103729303A (en) | Data writing and data reading methods of Flash | |
| CN103544110A (en) | Block-level continuous data protection method based on solid-state disc | |
| CN101630233B (en) | Data access method used for flash memory, storage system and controller | |
| CN103617101A (en) | Power fail safeguard method and device | |
| CN101567220A (en) | Bad block identifying method, storage system and controller thereof for flash memory | |
| CN101699476A (en) | Data processing method and device of smart cards | |
| CN101464836B (en) | Data memory method for flash memory device | |
| CN102890655A (en) | Memory storage device, memory controller and valid data recognition method thereof | |
| CN111090392A (en) | Cold and hot data separation method based on feature codes | |
| CN101441596B (en) | Method for improving flash memory medium read-write speed | |
| CN102521146B (en) | Data addressing storage method of flash memory block subdivision |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20110511 |