CN101488153A - Method for implementing high-capacity flash memory file system in embedded type Linux - Google Patents
Method for implementing high-capacity flash memory file system in embedded type Linux Download PDFInfo
- Publication number
- CN101488153A CN101488153A CNA2009100961243A CN200910096124A CN101488153A CN 101488153 A CN101488153 A CN 101488153A CN A2009100961243 A CNA2009100961243 A CN A2009100961243A CN 200910096124 A CN200910096124 A CN 200910096124A CN 101488153 A CN101488153 A CN 101488153A
- Authority
- CN
- China
- Prior art keywords
- page
- data
- section
- file system
- read
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a realizing method of an embedded Linux large capacity flash file system. The method comprises sectionalized storage, data page distribution, refuse collection, logic structure zoning, file system initialization, directory searching, file displacement location, data reading and data writing. The method realizes the embedded Linux large capacity Nand flash file system and is beneficial for the management of embedded Linux large capacity Nand flash. The realized large capacity Nand flash file system can be applied to the large capacity Nand flash management and is beneficial to the utilization of embedded storing.
Description
Technical field
The present invention relates to the embedded file system field, particularly relate to a kind of implementation method of high-capacity flash memory file system in embedded type Linux.
Background technology
Built-in Linux is a kind of small-scale operating system according to embedded OS requirement design.Form [6] by a Kernel (kernel) and system module that some customize according to demand.Its Kernel is very little, generally has only the size about hundreds of KB.Even add module and application program that other is necessary, shared storage space is also very little.
The principal feature of built-in Linux operating system has: simplifies the linux kernel of standard, is adapted to multiple CPU and hardware platform, and stable performance, reduction property is fine, and exploitation is used easily, can also use the application program on Unix or the Linux.Now, the multimedia application of built-in Linux under also can Windows.Built-in Linux operating system has the dirigibility of height, and the developer can customize or do suitable exploitation at an easy rate to it, satisfy the practical application needs.
Under built-in Linux, the effective ways of MOS and user data just are to use file system.The file system type that built-in Linux is supported is various, can select according to system storage media type and system applies purpose.
For the effective support to all kinds file system is provided, built-in Linux provides a Virtual File System layer, abstract user's file read-write operation, for the user, all file system system calls all have unified interface, and the specific implementation details all is hidden into Virtual File System in the calling of concrete file system.This Virtual File System layer has made things convenient for the transplanting of built-in Linux.
Along with updating and the continuous decline of manufacturing cost of Nand flash memory system fabrication technique, the Nand flash memory has replaced the main flow storage medium that other nonvolatile memorys become mobile storage market, and prospect is unlimited.These several years, many scientific research institutions of computerdom began or force Nand flash memory management Study on Technology both at home and abroad.Research direction is broadly divided into following:
1, Nand flash block mapping techniques: use a middle layer to drive, the Nand flash memory is modeled to the block device that a tradition can arbitrarily be read and write, and uses traditional disk file system to manage.FTL technology that American Ban you can well imagine out the fraction of the year of nineteen ninety-five and 1997 and NFTL technology are the forerunners of this technology.Current research focus comprises: focus discovery and processing, the scalable mapping techniques etc. of mass data storage in the Nand flash memory;
2, the realization research of the relevant complex data of the application on the Nand flash memory.As realizing Nand flash memory application on B-tree or R-tree, mobile communication and the sensing network or the like on the Nand flash memory.
Otherwise research comprises that also the Nand flash memory starts XIP technology and flash memory Study on real-time performance.
3, Nand flash memory dedicated file system research.A Nand flash memory file system must take into full account the physical characteristics of Nand flash memory, on the basis of effectively storage, and the serviceable life of having managed the Nand flash memory.About more the comprising of field of Nand flash memory file system research: file system fast initialization, file system journal management, system crash recovery, refuse collection and page distribution technique or the like.
If the research of present Nand flash memory file system lays stress on the storage characteristics of fully excavating the Nand flash memory, support erasable balance, the serviceable life of assurance Nand flash memory, lifting access performance by the appropriate design file system, but fully consider for high capacity Nand flash memory and file system shortage, particularly the random access to big file lacks effectively support.
The typical feature of high-capacity flash memory storer is exactly that the interior page number of storer has been compared significantly raising than normal flash.Therefore, the design high-capacity flash memory file system is except the problem of considering picture design normal flash file system, also need to consider and solve the new problem of being brought owing to the flash capacity increase, as: the storage of elongated, a large amount of page management method of thing followed file system initialization time, page or leaf allocation algorithm, large scale file and visit etc.Flash memory among the present invention promptly is meant jumbo Nand flash memory.ssss
Summary of the invention
The object of the present invention is to provide a kind of implementation method of high-capacity flash memory file system in embedded type Linux.
The technical scheme that the present invention solves its technical matters employing is as follows:
A kind of implementation method of high-capacity flash memory file system in embedded type Linux, the step of this method is as follows:
1) section storage:
Memory point all is to organize in the mode of section, and file data storing is in one section continuous page or leaf, and the interpolation of data and deletion also are that unit carries out with the section; Section is retractable, and has interpolation, deletes, merges, cuts apart; All sections leave in the B-tree;
2) data page distributes:
When distributing the blank page, the page of catalogue is opened in the different pieces with the page of data branch, avoided " all copies " like this, thereby reduced the copy number of times of the page, improved efficient;
Every type piece has and only has a piece to be in " current distribution " state.New page order from the current piece that distributes is distributed, if current block is full, then order is sought next free block;
3) refuse collection:
For dirty data, the algorithm that uses multiple existing recovery method to mix carries out garbage reclamation, with greedy method and at random system of selection mix use in proportion: when satisfying specific small probability condition, this specific small probability condition is set according to different applied environments when specific implementation, and Garbage Collector can attempt to use callable page of system of selection at random; And in other cases, then use greedy method to reclaim the piece of " dirty ";
4) zoned logic structure:
Represent that at first section of subregion flash block deposits the index node section of file system, the data representation that the page is deposited in the piece be the piece number of some pieces in the subregion; These pieces are being deposited storage object head element information, and each storage object head element distributes a page; This section begins to distribute from subregion, and fixing length.Concrete length is seen according to the size of subregion and is set; Last part is a data segment, and it comprises effective page of data, the dirty data page and the blank page;
5) file system initialization:
1, at first wants initialization files system correlated variables;
2, initialization Nand flash memory, and set up the memory device object and manage.
3, scan N and flash partition reads the piece that each has file and directory index node successively from the index node section that subregion begins;
4, utilize an object page to set up the initialization information of object;
6) directory search:
From root directory or current directory zone-by-zone analysis filename, up to finally finding this document;
7) file displacement location:
Displacement location is one and finally finds a leafy node process from root node, at first obtains parameter and carry out integrity checking; If qualified, then from root node, find corresponding child node, if this child node and the former all not enough file displacement of byte number of child node then need to seek next child node, up to finding corresponding child node according to given displacement; After determining child node, from the leaf nodes of child node, search successively again, up to the section that finds correspondence, and the local displacement in definite section, and return; If do not have next child node or corresponding leaf node, and determine the displacement on the determined section of this leaf node;
8) data read:
Flow process from HNFS file system reading of data:
1, at first obtains side-play amount that reads and the byte number that reads;
2, determine section and section intrinsic displacement;
3, begin to read circulation, cycling condition is set reads for also having data;
4, from this section begin the order read each the page or leaf with copying data to buffer, revise buffer pointer offset and remaining data amount simultaneously and read skew;
5, read a page, at first check in the system page buffer memory whether corresponding cache is arranged,, then read in caching of page from the Nand flash memory if do not have;
If 6 certain sector data have run through, then select next next-door neighbour's section to read, up to reading enough byte numbers, if file does not have enough byte numbers;
7, the byte number that returns buffer and read altogether after finishing;
9) data write:
Write the flow process of data to the HNFS file system:
1, the byte number that at first obtains the document misregistration amount and write;
2, determine section and section intrinsic displacement;
3, begin to write circulation, cycling condition is set writes for also having data;
4, read a page earlier, at first check in the system page buffer memory whether corresponding cache is arranged,, then read in caching of page from the Nand flash memory if do not have;
5, from buffer, copy the data order to this section internal displacement and begin, revise buffer pointer offset and remaining data amount simultaneously and read skew;
If 6 take the write through method, then directly select a free page that the buffer memory page is write, write the successfully corresponding stored section of back corresponding modify object head.If mistake in writing process, occurs, then report an error;
7, the data that write have not also been write, and then order determines that next next-door neighbour's section writes, until all having write.Call the dirty page in the garbage collector recovery system when system's free page is not enough;
8, return the byte number that writes after finishing.
The present invention compares with background technology, and the useful effect that has is:
The present invention is a kind of implementation method of high-capacity flash memory file system in embedded type Linux, its major function is at the storage characteristics of high capacity Nand flash memory, the thought of data segments storage has been proposed, with the method that realizes high capacity Nand flash memory file system, and describe the process of system initialization, directory search, displacement location and the reading and writing data etc. of high capacity Nand flash memory file system HNFS in detail.This method has realized high capacity Nand flash memory file system under built-in Linux, helps the management of high capacity Nand flash memory under the built-in Linux.
(1) high efficiency.This method has realized high capacity Nand flash memory file system under built-in Linux, can manage high capacity Nand flash memory efficiently.
(2) practicality.Realization to the file system of high capacity Nand flash memory helps the utilization to embedded storage.
Description of drawings
Fig. 1 is an implementation process synoptic diagram of the present invention;
Fig. 2 is a file system initialization process flow diagram of the present invention;
Fig. 3 is a file system files displacement location process flow diagram of the present invention;
Fig. 4 is that file system data of the present invention reads process flow diagram;
Fig. 5 is that file system data of the present invention writes process flow diagram.
Embodiment
The present invention is a kind of implementation method of high-capacity flash memory file system in embedded type Linux, below in conjunction with Fig. 1 its specific implementation process is described.
1) section storage:
The maximum characteristics of high capacity Nand flash memory are exactly that data page is many, and memory capacity is big.Therefore when carrying out design Storage, need design to support the storage and the visit of large scale file, consider the many situations of small documents in the file system simultaneously.And storage space saving and File Compress are not main storage target.
The section storage administration is different from the piece that often is indicated in disk file system, the all memory points of section storage administration all are to organize in the mode of section, file data leaves in one section continuous page or leaf as much as possible, and the interpolation of data and deletion also are that unit carries out with the section; Section is telescopic, and has interpolation, deletion, merging, dividing function.All sections leave in the B-tree.For the small documents of frequent modification, carry out that data are write fashionable needs corresponding upgating object head element and data page gets final product.For big file and the super large file that high-frequency reads at random, section management can carry out displacement calculating to compare file system faster speed such as YAFFS.The file system operating characteristic of above mentioning can be effectively satisfied in the section management, reduces excess loss, raising speed of random access that indirect index brings.
In the present invention, file, hard link and catalogue are abstracted into storage object, and each object has an object head, deposit the information of the aspect such as control, visit of object, an object head of each the object head specific assigned page.Be called high capacity Nand flash memory file system HNFS according to the file system that the inventive method realized, hereinafter to be referred as HNFS.
Catalogue has higher access frequency than file.If the object head is placed in independent one page, the renewal frequency of object head is higher, and this page can repeatedly use so, deposits a plurality of versions of storage object element, has both helped reducing the loss of Nand flash memory, also is convenient to recover from possible system crash.And if with 2 or the shared page of more object head element, any file modifying must cause a modification of the element page so, the renewal frequency of this page is just too high like this, has run counter to the saving space, reduces the original intention of Nand flash memory loss; If data and an element are placed in the page, because the access frequency difference of common catalogue and ordinary file data is bigger, this also brings data often to shift without reason, and renewal of the page is problem faster.
According to the storage characteristics of Nand flash memory, improve the rate of loading of file system simultaneously.The beginning and reserve several pieces in advance of flash partition, as file inode piece index.When loading high-capacity flash memory file system, the position of reading the index node piece from these pages is read index node information again from the index node piece, thereby makes up whole file system in internal memory.
2) data page distributes:
Different with disk, the Nand flash memory does not exist to be sought rail and postpones, so has higher access speed during reading of data.But because the Nand flash memory can only write data to the blank page, so the overall performance of writing speed and blank page distribution method, storage space situation and garbage collector performance and file system is closely related.Distribution method is as follows: in certain piece, blank page is according to the order assignment that increases progressively, the page in piece distribute be over after, select a blank block to participate in the distribution again in addition.In addition, also to be used for " all copies " for refuse collection keeps 2-3 pieces.When not having the blank page in the system, select one the dirty erasable blank block that becomes to use.
From the angle of file system, the page of storage system metadata such as catalog page just have higher turnover rate than general data face.Therefore, when distributing the blank page, the page and the page of data branch of catalogue is opened in the different piece (Block), avoided " all copies " (Full copy) like this, thereby reduced the copy number of times of the page, improved efficient.In the present invention, every type piece has and only has a piece to be in " current distribution " state.New page order from the current piece that distributes is distributed, if current block is full, then order is sought next free block.
3) refuse collection:
When the storage space empty page is lower than a predetermined value and is, need to use garbage collector to reclaim and bury page of data, return to system.
For dirty data, the algorithm that uses multiple existing recovery method to mix carries out garbage reclamation, greedy method is mixed use by a certain percentage with system of selection at random: when satisfying specific small probability condition (this specific small probability condition in specific implementation time set according to different applied environments), Garbage Collector can be attempted with callable page of system of selection at random; And in other cases, then use greedy method to reclaim the piece of " dirty ".
When reclaiming the dirty data piece, treat dissimilar dirty data pieces and also adopt diverse ways.If dirty that reclaims is used for storing directory, so just the effective page in the piece is copied in the piece of other storing directory and go, if what dirty of reclaiming deposited is the ordinary file data, so just the effective page in the piece is copied in other pieces of depositing the ordinary file data and go.
4) zoned logic structure:
Represent that at first section of subregion flash block deposits the index node section of file system (Inode-StoredBlock of Segment), the data representation that the page is deposited in the piece be the piece number of some pieces in the subregion.These pieces are being deposited storage object head element information, and each storage object head element distributes a page.This section begins to distribute from subregion, and fixing length.Concrete length is seen according to the size of subregion and is set.Last part is a data segment, and it comprises effective page of data, the dirty data page and the blank page.
When system initialization,, finish file system and load by read head page info in the index node section.
5) file system initialization:
The file system initialization process, as shown in Figure 2:
1, at first wants initialization files system correlated variables, as superb, directory tree, target cache or the like.
2, initialization Nand flash memory, and set up the memory device object and manage.
3, scan N and flash partition, from the index node section that subregion begins, read the piece that each has file and directory index node successively, in each page of piece, read again and have a piece of page piece number, some pieces in each piece number sensing subregion are represented with no symbol 16 bits of two bytes.Each page in these data blocks is being deposited the header of catalogue and file object.
4, utilize an object page to set up the initialization information of object, this comprises access control information, status information and canned data.Each storage object is by the structure management storage of a class B-tree.Each object also can be added in the directory tree.
Along with constantly reading in of index node, file system being initialised gradually.In HNFS, all subobjects under the catalogue link together by pointer, and file data manages by the B-tree.To set up one at last be root node to the system to each file from root directory is the search tree of leaf node.
6) directory search:
The path of a Linux filename is that a series of directory names that separated by "/" are formed, and the name with file finishes at last.For example a filename is/home/ruby/.cshrc, here/home and/ruby is a dirname, the name of file is .cshrc.As the Unix system of all other, Linux does not stress to the form of filename itself.The maximum length that it is supported is 1024 characters, is made up of printable character.In order to find the memory object of certain file correspondence, the file system that the inventive method realized must be from root directory or current directory zone-by-zone analysis filename, up to finally finding this document.
First that needs is to liking the object of file system root (root).Can in the superblock of file system, obtain its value.In order to read a HNFS object, must in its chained list of parent directory, find it.For example, must find home earlier, and then from the subobject chained list of home, find ruby for ruby.
Home is in the subobject chained list among ＂/＂.From its inlet item "/", can obtain the pointer of this object.Must find ruby by this directory object (at first obtain the pointer that points to this object, from the subobject chained list of this object, obtain the pointer of the ruby of sensing then).From the data that obtain, obtain/inlet of the number of home/ruby catalogue memory object.Read at last and point to the memory object data of describing catalogue/home/ruby.And from the data block of its sensing, find the memory object numerical value of .csshrc.From this memory object, can locate the data block that comprises this document data.
7) file displacement location:
In HNFS, the purpose of file displacement location is exactly according to the given displacement of user, the storage data segment of specified data and the displacement in data segment.
As previously mentioned, in HNFS, the memory file data are with the structure organization of B-tree.The pointer of leaf node and the byte number of this node storage are pointed in tree Zhi Jiedian storage in the B-tree.And each leaf node is deposited a pointer that points to storage compartments and the byte number of depositing in this section.The B-tree is set up when file system initialization.
As shown in Figure 3, displacement location be one from root node, finally find a leafy node process.At first obtain parameter and carry out integrity checking.If qualified, then from root node, find corresponding child node, if this child node and the former all not enough file displacement of byte number of child node then need to seek next child node, up to finding corresponding child node according to given displacement.After determining child node, from the leaf nodes of child node, search successively again, up to the section that finds correspondence, and the local displacement in definite section, and return.If do not have next child node or corresponding leaf node, and determine the displacement on the determined section of this leaf node.
8) data read:
From the flow process of HNFS file system reading of data as shown in Figure 4:
1, at first obtains side-play amount that reads and the byte number that reads;
2, determine section and section intrinsic displacement;
3, begin to read circulation, cycling condition is set reads for also having data;
4, from this section begin the order read each the page or leaf with copying data to buffer, revise buffer pointer offset and remaining data amount simultaneously and read skew;
5, read a page, at first check in the system page buffer memory whether corresponding cache is arranged,, then read in caching of page from the Nand flash memory if do not have;
If 6 certain sector data have run through, then select next next-door neighbour's section to read, up to reading enough byte numbers, if file does not have enough byte numbers;
7, the byte number that returns buffer and read altogether after finishing.
9) data write:
The flow process that writes data to the HNFS file system is as shown in Figure 5:
1, the byte number that at first obtains the document misregistration amount and write;
2, determine section and section intrinsic displacement;
3, begin to write circulation, cycling condition is set writes for also having data;
4, read a page earlier, at first check in the system page buffer memory whether corresponding cache is arranged,, then read in caching of page from the Nand flash memory if do not have;
5, from buffer, copy the data order to this section internal displacement and begin, revise buffer pointer offset and remaining data amount simultaneously and read skew;
If 6 take the write through method, then directly select a free page that the buffer memory page is write, write the successfully corresponding stored section of back corresponding modify object head.If mistake in writing process, occurs, then report an error.
7, the data that write have not also been write, and then order determines that next next-door neighbour's section writes, until all having write.Call the dirty page in the garbage collector recovery system when system's free page is not enough;
8, return the byte number that writes after finishing.
Claims (1)
1. the implementation method of a high-capacity flash memory file system in embedded type Linux is characterized in that the step of this method is as follows:
1) section storage:
Memory point all is to organize in the mode of section, and file data storing is in one section continuous page or leaf, and the interpolation of data and deletion also are that unit carries out with the section; Section is retractable, and has interpolation, deletes, merges, cuts apart; All sections leave in the B-tree;
2) data page distributes:
When distributing the blank page, the page of catalogue is opened in the different pieces with the page of data branch, avoided " all copies " like this, thereby reduced the copy number of times of the page, improved efficient;
Every type piece has and only has a piece to be in " current distribution " state.New page order from the current piece that distributes is distributed, if current block is full, then order is sought next free block;
3) refuse collection:
For dirty data, the algorithm that uses multiple existing recovery method to mix carries out garbage reclamation, with greedy method and at random system of selection mix use in proportion: when satisfying specific small probability condition, this specific small probability condition is set according to different applied environments when specific implementation, and Garbage Collector can attempt to use callable page of system of selection at random; And in other cases, then use greedy method to reclaim the piece of " dirty ";
4) zoned logic structure:
Represent that at first section of subregion flash block deposits the index node section of file system, the data representation that the page is deposited in the piece be the piece number of some pieces in the subregion; These pieces are being deposited storage object head element information, and each storage object head element distributes a page; This section begins to distribute from subregion, and fixing length.Concrete length is seen according to the size of subregion and is set; Last part is a data segment, and it comprises effective page of data, the dirty data page and the blank page;
5) file system initialization:
1, at first wants initialization files system correlated variables;
2, initialization Nand flash memory, and set up the memory device object and manage.
3, scan N and flash partition reads the piece that each has file and directory index node successively from the index node section that subregion begins;
4, utilize an object page to set up the initialization information of object;
6) directory search:
From root directory or current directory zone-by-zone analysis filename, up to finally finding this document;
7) file displacement location:
Displacement location is one and finally finds a leafy node process from root node, at first obtains parameter and carry out integrity checking; If qualified, then from root node, find corresponding child node, if this child node and the former all not enough file displacement of byte number of child node then need to seek next child node, up to finding corresponding child node according to given displacement; After determining child node, from the leaf nodes of child node, search successively again, up to the section that finds correspondence, and the local displacement in definite section, and return; If do not have next child node or corresponding leaf node, and determine the displacement on the determined section of this leaf node;
8) data read:
Flow process from HNFS file system reading of data:
1, at first obtains side-play amount that reads and the byte number that reads;
2, determine section and section intrinsic displacement;
3, begin to read circulation, cycling condition is set reads for also having data;
4, from this section begin the order read each the page or leaf with copying data to buffer, revise buffer pointer offset and remaining data amount simultaneously and read skew;
5, read a page, at first check in the system page buffer memory whether corresponding cache is arranged,, then read in caching of page from the Nand flash memory if do not have;
If 6 certain sector data have run through, then select next next-door neighbour's section to read, up to reading enough byte numbers, if file does not have enough byte numbers;
7, the byte number that returns buffer and read altogether after finishing;
9) data write:
Write the flow process of data to the HNFS file system:
1, the byte number that at first obtains the document misregistration amount and write;
2, determine section and section intrinsic displacement;
3, begin to write circulation, cycling condition is set writes for also having data;
4, read a page earlier, at first check in the system page buffer memory whether corresponding cache is arranged,, then read in caching of page from the Nand flash memory if do not have;
5, from buffer, copy the data order to this section internal displacement and begin, revise buffer pointer offset and remaining data amount simultaneously and read skew;
If 6 take the write through method, then directly select a free page that the buffer memory page is write, write the successfully corresponding stored section of back corresponding modify object head.If mistake in writing process, occurs, then report an error;
7, the data that write have not also been write, and then order determines that next next-door neighbour's section writes, until all having write.Call the dirty page in the garbage collector recovery system when system's free page is not enough;
8, return the byte number that writes after finishing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2009100961243A CN101488153A (en) | 2009-02-12 | 2009-02-12 | Method for implementing high-capacity flash memory file system in embedded type Linux |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNA2009100961243A CN101488153A (en) | 2009-02-12 | 2009-02-12 | Method for implementing high-capacity flash memory file system in embedded type Linux |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN101488153A true CN101488153A (en) | 2009-07-22 |
Family
ID=40891043
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNA2009100961243A Pending CN101488153A (en) | 2009-02-12 | 2009-02-12 | Method for implementing high-capacity flash memory file system in embedded type Linux |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101488153A (en) |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101825995A (en) * | 2010-05-11 | 2010-09-08 | 太仓市同维电子有限公司 | Method for supporting reading and writing of files on small flash memory partitions |
| CN102023992A (en) * | 2009-09-18 | 2011-04-20 | 青岛海信电器股份有限公司 | Content browsing method and apparatus |
| CN102253894A (en) * | 2011-06-14 | 2011-11-23 | 沈阳建筑大学 | Method for file storage in continuous space of flash memory in equal probability manner |
| CN102323907A (en) * | 2011-08-19 | 2012-01-18 | 长安大学 | Method for embedded ARM (advanced RISC machines) processor to store and delete NANDFLASH data |
| CN101789004B (en) * | 2010-01-22 | 2012-07-18 | 北京北大众志微系统科技有限责任公司 | Rapid-start method and device of Flash file system |
| CN102135971B (en) * | 2010-01-25 | 2013-03-06 | 英业达股份有限公司 | File management method for server |
| CN103853850A (en) * | 2014-03-28 | 2014-06-11 | 北京工业大学 | Method and system for eliminating data for user-centered Android system |
| CN103870364A (en) * | 2014-03-31 | 2014-06-18 | 北京工业大学 | YAFFS2 file final-version recovery method based on timestamp |
| CN107168886A (en) * | 2016-03-07 | 2017-09-15 | 爱思开海力士有限公司 | Data storage device and its operating method |
| CN107533506A (en) * | 2015-05-06 | 2018-01-02 | 华为技术有限公司 | Rubbish recovering method and equipment in log-structuredization file system |
| CN108628753A (en) * | 2017-03-24 | 2018-10-09 | 华为技术有限公司 | Memory headroom management method and device |
| CN109656874A (en) * | 2018-11-28 | 2019-04-19 | 烟台蓝洋电子科技有限责任公司 | Binary file management system implementation method |
| CN109800185A (en) * | 2018-12-29 | 2019-05-24 | 上海霄云信息科技有限公司 | A kind of data cache method in data-storage system |
| CN110554842A (en) * | 2019-09-06 | 2019-12-10 | 华大半导体有限公司 | Method for simulating EEPROM (electrically erasable programmable read-Only memory) by using flash memory |
| CN112231246A (en) * | 2020-10-31 | 2021-01-15 | 王志平 | Method for realizing processor cache structure |
| CN112527745A (en) * | 2020-12-18 | 2021-03-19 | 厦门市美亚柏科信息股份有限公司 | Embedded file system multi-partition analysis method, terminal device and storage medium |
| CN113157395A (en) * | 2021-04-20 | 2021-07-23 | 上海泓戟信息科技有限公司 | Method for optimizing computer memory data synchronization |
| CN113485974A (en) * | 2021-06-02 | 2021-10-08 | 翱捷科技股份有限公司 | Method and device for using YAFFS (flash architecture) for NOR flash memory |
| CN113742127A (en) * | 2021-09-16 | 2021-12-03 | 重庆大学 | Fault recovery method for bare flash memory file system |
| CN114090618A (en) * | 2022-01-18 | 2022-02-25 | 江西树格科技有限公司 | Data reading method, system, computer and readable storage medium |
-
2009
- 2009-02-12 CN CNA2009100961243A patent/CN101488153A/en active Pending
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102023992A (en) * | 2009-09-18 | 2011-04-20 | 青岛海信电器股份有限公司 | Content browsing method and apparatus |
| CN101789004B (en) * | 2010-01-22 | 2012-07-18 | 北京北大众志微系统科技有限责任公司 | Rapid-start method and device of Flash file system |
| CN102135971B (en) * | 2010-01-25 | 2013-03-06 | 英业达股份有限公司 | File management method for server |
| CN101825995B (en) * | 2010-05-11 | 2011-08-03 | 太仓市同维电子有限公司 | Method for supporting reading and writing of files on small flash memory partitions |
| CN101825995A (en) * | 2010-05-11 | 2010-09-08 | 太仓市同维电子有限公司 | Method for supporting reading and writing of files on small flash memory partitions |
| CN102253894A (en) * | 2011-06-14 | 2011-11-23 | 沈阳建筑大学 | Method for file storage in continuous space of flash memory in equal probability manner |
| CN102253894B (en) * | 2011-06-14 | 2013-09-04 | 沈阳建筑大学 | Method for file storage in continuous space of flash memory in equal probability manner |
| CN102323907A (en) * | 2011-08-19 | 2012-01-18 | 长安大学 | Method for embedded ARM (advanced RISC machines) processor to store and delete NANDFLASH data |
| CN103853850A (en) * | 2014-03-28 | 2014-06-11 | 北京工业大学 | Method and system for eliminating data for user-centered Android system |
| CN103853850B (en) * | 2014-03-28 | 2017-02-15 | 北京工业大学 | Method and system for eliminating data for user-centered Android system |
| CN103870364A (en) * | 2014-03-31 | 2014-06-18 | 北京工业大学 | YAFFS2 file final-version recovery method based on timestamp |
| CN103870364B (en) * | 2014-03-31 | 2017-10-24 | 北京工业大学 | A kind of final version restoration methods of YAFFS2 files based on timestamp |
| CN107533506B (en) * | 2015-05-06 | 2021-03-23 | 华为技术有限公司 | Garbage collection method and equipment in log structured file system |
| CN107533506A (en) * | 2015-05-06 | 2018-01-02 | 华为技术有限公司 | Rubbish recovering method and equipment in log-structuredization file system |
| CN107168886B (en) * | 2016-03-07 | 2021-06-22 | 爱思开海力士有限公司 | Data storage device and operation method thereof |
| CN107168886A (en) * | 2016-03-07 | 2017-09-15 | 爱思开海力士有限公司 | Data storage device and its operating method |
| CN108628753A (en) * | 2017-03-24 | 2018-10-09 | 华为技术有限公司 | Memory headroom management method and device |
| CN109656874A (en) * | 2018-11-28 | 2019-04-19 | 烟台蓝洋电子科技有限责任公司 | Binary file management system implementation method |
| CN109656874B (en) * | 2018-11-28 | 2024-03-08 | 山东蓝洋智能科技有限公司 | Method for implementing file management system in dual system |
| CN109800185B (en) * | 2018-12-29 | 2023-10-20 | 上海霄云信息科技有限公司 | Data caching method in data storage system |
| CN109800185A (en) * | 2018-12-29 | 2019-05-24 | 上海霄云信息科技有限公司 | A kind of data cache method in data-storage system |
| CN110554842A (en) * | 2019-09-06 | 2019-12-10 | 华大半导体有限公司 | Method for simulating EEPROM (electrically erasable programmable read-Only memory) by using flash memory |
| CN112231246A (en) * | 2020-10-31 | 2021-01-15 | 王志平 | Method for realizing processor cache structure |
| CN112527745B (en) * | 2020-12-18 | 2022-07-12 | 厦门市美亚柏科信息股份有限公司 | Embedded file system multi-partition analysis method, terminal device and storage medium |
| CN112527745A (en) * | 2020-12-18 | 2021-03-19 | 厦门市美亚柏科信息股份有限公司 | Embedded file system multi-partition analysis method, terminal device and storage medium |
| CN113157395A (en) * | 2021-04-20 | 2021-07-23 | 上海泓戟信息科技有限公司 | Method for optimizing computer memory data synchronization |
| CN113485974A (en) * | 2021-06-02 | 2021-10-08 | 翱捷科技股份有限公司 | Method and device for using YAFFS (flash architecture) for NOR flash memory |
| CN113485974B (en) * | 2021-06-02 | 2023-08-01 | 翱捷科技股份有限公司 | Method and device for using YAFFS (gamma ray source flash memory) |
| CN113742127A (en) * | 2021-09-16 | 2021-12-03 | 重庆大学 | Fault recovery method for bare flash memory file system |
| CN114090618A (en) * | 2022-01-18 | 2022-02-25 | 江西树格科技有限公司 | Data reading method, system, computer and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101488153A (en) | Method for implementing high-capacity flash memory file system in embedded type Linux | |
| CN107180092B (en) | File system control method and device and terminal | |
| Chang et al. | An efficient management scheme for large-scale flash-memory storage systems | |
| CN101187901B (en) | High speed cache system and method for implementing file access | |
| CN100483420C (en) | Fine grit document and catalogs version management method based on snapshot | |
| CN108628753B (en) | Memory space management method and device | |
| CN101980177B (en) | Method and device for operating Flash | |
| US20110258374A1 (en) | Method for optimizing the memory usage and performance of data deduplication storage systems | |
| CN103838853B (en) | Mixed file system based on different storage media | |
| CN100424699C (en) | Attribute extensible object file system | |
| CN105045850B (en) | Junk data recovery method in cloud storage log file system | |
| Sarwat et al. | Generic and efficient framework for search trees on flash memory storage systems | |
| CN115427941A (en) | Data management system and control method | |
| WO2021208239A1 (en) | Low-latency file system address space management method and system, and medium | |
| CN102253985B (en) | File system data management method and system | |
| CN102024060B (en) | Method for memorizing files and restoring error deleted files | |
| CN113704217A (en) | Metadata and data organization architecture method in distributed persistent memory file system | |
| Jin et al. | Exploring index structures for zoned namespaces SSDs | |
| Carniel et al. | A generic and efficient framework for flash-aware spatial indexing | |
| CN114676072A (en) | Data processing method and device | |
| CN105630410B (en) | The method for managing file storage area using non-volatile random read-write storage chip dynamic | |
| KR20140021793A (en) | Database method for b+ tree based on pram | |
| KR100907477B1 (en) | Apparatus and method for managing index of data stored in flash memory | |
| Lee et al. | Efficient logging of metadata using NVRAM for NAND flash based file system | |
| WO2022262381A1 (en) | Data compression method and apparatus |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20090722 |