CN103838853A - Mixed file system based on different storage media - Google Patents
Mixed file system based on different storage media Download PDFInfo
- Publication number
- CN103838853A CN103838853A CN201410096969.3A CN201410096969A CN103838853A CN 103838853 A CN103838853 A CN 103838853A CN 201410096969 A CN201410096969 A CN 201410096969A CN 103838853 A CN103838853 A CN 103838853A
- Authority
- CN
- China
- Prior art keywords
- file system
- addressing
- metadata
- medium
- volatile
- 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
Images
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
- G06F3/0668—Interfaces specially adapted for storage systems adopting a particular infrastructure
- G06F3/0671—In-line storage system
- G06F3/0683—Plurality of storage devices
- G06F3/0685—Hybrid storage combining heterogeneous device types, e.g. hierarchical storage, hybrid arrays
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/10—File systems; File servers
- G06F16/18—File system types
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a mixed file system based on different storage media. The storage media comprise non-volatile small-granularity addressing media and non-volatile large-granularity addressing media, and the two kinds of media are addressed according to the unified sequence. The non-volatile small-granularity addressing media are used for storing metadata of the file system, and the non-volatile large-granularity addressing media are used for storing data of the file system. The file system calls different granularities of reading or writing for different address ranges under different situations. According to the mixed file system, large data volume reading and writing and small data volume reading and writing are allocated to the corresponding media according to different situations, the reading and writing speed is improved, and the storing efficiency of the file system is improved. The storage space is fully utilized, superfluous redundancy does not exist, additional I/O is reduced, and reading and writing performance is improved. The mixed file system has the high reliability, expandability and I/O speed.
Description
Technical field
The invention belongs to computer memory system technical field, more specifically, relate to a kind of mixed file system based on different storage mediums.
Background technology
In Computer Storage subsystem, due to the read and write access time delay difference of different storage mediums, in Computer Architecture, usually provide multi-level memory hierarchy.Its access time of storage medium that memory capacity is larger is longer.At present, normally adopt magnetic storage technology to realize jumbo storage, its read-write postpones large, and read-write granularity is large.File system access during as the large capacity storage of peripheral hardware, can consume a lot of time, reduces overall performance.
For improving performance, field of storage has been done a lot of research.The appearance of solid-state disk has caused a tide especially.In in recent years, a series of mixing storage scheme is suggested.Wherein the scheme of main flow has two kinds: as shown in Fig. 1 (a), a kind of scheme adopts reads and writes the buffer memory of fast medium as the slow medium of read-write, such as adopting solid-state disk as buffer memory, be the buffer memory of solid-state disk as disk, by often access data buffer storage in solid-state disk, for improving the hit rate of solid-state disk in this storage scheme, in succession there is the strategy of series of optimum buffer memory; As shown in Fig. 1 (b), another kind of scheme adopts reads and writes the fast medium medium slow with read-write as mixing storage, take solid-state disk and hard disk as example, what data were unique is stored in solid-state disk or hard disk, and data exist the selection algorithm in solid-state disk or hard disk to lay particular stress on a difference according to performance also to be had multiple.In all schemes, owing to having introduced solid-state disk (Solid State Disk, SSD), in the speed of file reading data, improve to some extent, but not high for the access performance optimization of writing and frequently revise, particularly upgrade for irregular small documents.Revise metadata because spend a large amount of time, also due to device itself factor---the erasable characteristic of SSD, has a large amount of additionally writing.
Summary of the invention
In order to improve the problem of additionally writing in prior art, raising modification metadata and the more speed of new data, the present invention proposes a kind of new mixing storage organization, using the medium of non-volatile small grain size addressing with the medium of non-volatile coarsegrain addressing as mixing storage, and for this mixing storage organization, new file system has been proposed.
To achieve these goals, the invention provides a kind of mixed file system based on different storage mediums, described storage medium comprises the medium of non-volatile small grain size addressing and the medium of non-volatile coarsegrain addressing, and two media is unified sequential addressing; Wherein, the medium of described non-volatile small grain size addressing is for the metadata of storing documents system, and the medium of described non-volatile coarsegrain addressing is for the data of storing documents system; Described file system is called varigrained reading or writing in different situations to different address realms.
As a further improvement on the present invention, described two media is unified sequential addressing and is specially: the medium of described non-volatile small grain size addressing is addressed in LAP Low Address Part, and the medium of described non-volatile coarsegrain addressing is addressed in high address section.
Particularly, the metadata of described file system refers to the data structure in managed storage space, metadata with file in file system, and directory tree structure, specifically comprise attribute and the content of the attribute of data structure, file of data structure, the management allocation of space of managing whole file system and pointer, catalogue.
Particularly, the data of described file system refer to the content of the file that leaves data block district in.
As a further improvement on the present invention, described file system is called varigrained reading or writing in different situations to different address realms and is comprised: the metadata that needs buffer memory part at initial phase, and the metadata of run duration buffer memory is need to replace time, adopts reading of coarsegrain; In service in file system, the content of access file or catalogue, adopts reading of coarsegrain.
As a further improvement on the present invention, described file system is called varigrained reading or writing in different situations to different address realms and is comprised: in service in file system, for searching of file or catalogue, when the metadata amount of needs access is during in byte number magnitude, adopt small grain size to read.
As a further improvement on the present invention, described file system is called varigrained reading or writing in different situations to different address realms and is comprised: in service in file system, when needs are more when new metadata, metadata on the medium of non-volatile small grain size addressing is adopted to writing back of small grain size, wherein need the more situation of new metadata to comprise the deletion to file or catalogue, or create, or revise, or rename.
As a further improvement on the present invention, described file system is called varigrained reading or writing in different situations to different address realms and is comprised: in service in file system, in the time of needs batch updating data, to data acquisition the writing back with coarsegrain on the medium of non-volatile coarsegrain addressing, wherein need the more situation of new data to comprise the deletion to file, or create, or revise.
Preferably, the medium of described non-volatile small grain size addressing is phase transition storage, and the medium of described non-volatile coarsegrain addressing is solid-state disk.
Particularly, if described file system is EXT2 form, the metadata of described file system comprises the content of superblock, piece group descriptor, inode bitmap, data block bitmap, inode table and catalogue.
The above technical scheme of conceiving by the present invention compared with prior art, has following beneficial effect:
1, the present invention is by adopting the medium of non-volatile small grain size addressing and the medium composition hybrid storage medium of non-volatile coarsegrain addressing, the metadata of storing documents system in the medium of non-volatile small grain size addressing, the data of storing documents system in the medium of non-volatile coarsegrain addressing, and call varigrained reading or writing for different situations and different address realms, thereby the read-write of the read-write of big data quantity and small data quantity is distributed on corresponding medium by different situations, improve read or write speed, promoted the storage efficiency of file system;
2, the present invention is left metadata on the medium of non-volatile small grain size addressing, renewal for metadata adopts writing of small grain size, thereby in the time of new metadata more, do not need whole data block to write back, thereby reduced the data volume writing back, avoided the problem of additionally writing;
3, the present invention is in service in file system, for searching of file or catalogue, when the metadata amount of needs access is during in byte number magnitude, adopt small grain size to read, thereby in the time reading metadata, do not need whole data block to read, thereby reduced reading of additional data volume;
4, because its read-write delay of medium of small grain size addressing is little, reduce the access time of metadata, promoted the read-write efficiency to metadata of file system;
5, in general, the method for described mixing storage provided by the invention and current method are compared, and its utilization factor is high, and read or write speed is faster, can reduce I/O read-write amount, have high reliability.
Accompanying drawing explanation
Fig. 1 is the typical storage architecture that mixes in prior art;
Fig. 2 is a kind of typical mixed architecture based on different storage mediums in the present invention;
Fig. 3 is a kind of typical file system physical layout in the embodiment of the present invention;
Fig. 4 is the read operation schematic diagram of file system in the embodiment of the present invention;
Fig. 5 is the write operation schematic diagram of file system in the embodiment of the present invention.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearer, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein, only in order to explain the present invention, is not intended to limit the present invention.In addition,, in each embodiment of described the present invention, involved technical characterictic just can combine mutually as long as do not form each other conflict.
First, some nouns and term in the present invention are made an explanation:
Coarsegrain and small grain size: coarsegrain is generally the size of page or piece; Small grain size is general to be more or less the same with the structure of a metadata, such as being 128Byte, much smaller than the size of page or piece; The read-write of coarsegrain can be identical with the read-write granularity of the medium of non-volatile coarsegrain addressing or be greater than its read-write granularity, and the read-write granularity of the read-write of small grain size and the medium of non-volatile small grain size addressing is identical or being greater than it reads and writes granularity.
File system metadata: the metadata of file system refers to metadata and the directory tree structure of file in the management data structure in space and file system here, comprises attribute and the content etc. of manage the attribute of the data structure of whole file system, the data structure of managing allocation of space, file and pointer, catalogue; Take EXT2 as example, the content of superblock, piece group descriptor, inode bitmap, data block bitmap, inode table and catalogue all belongs to the metadata of file system;
The data of file system: the content of file; Take EXT2 as example, the data of file system are the contents that leaves the file in data block district in.
For improving the problem of additionally writing, raising modification metadata and the more speed of new data, a kind of new mixing storage organization has been proposed in the present invention, using the medium of non-volatile small grain size addressing with the medium of non-volatile coarsegrain addressing as mixing storage, and for this mixing storage organization, new file system has been proposed.
A kind of new file system based on above-mentioned mixing storage provided by the invention.In this new mixing storage, metadata is placed on the medium of non-volatile small grain size addressing, and data is placed on the medium of non-volatile coarsegrain addressing.Two media is unified sequential addressing, and file system has different read-write interfaces to different address spaces different in the situation that.
File system can be called four class read-writes, is respectively that small grain size is read, small grain size is write, coarsegrain is read, coarsegrain is write, and file system can be called varigrained read-write in different situations to different address realms.Small grain size is general to be more or less the same with the structure of a metadata, and such as being 128Byte, much smaller than the size of page or piece, coarsegrain is generally the size of page or piece.
This mixed file system embodiment as follows:
In building process, two media is unified to sequential addressing, the medium of non-volatile small grain size addressing is at LAP Low Address Part, and the medium of non-volatile coarsegrain addressing is in high address section;
In layout, the metadata of the address space storing documents system of the medium of non-volatile small grain size addressing, the data of the address space storing documents system of the medium of non-volatile coarsegrain addressing;
The metadata of file system refers to metadata and the directory tree structure of file in the management data structure in space and file system here, comprises attribute and the content etc. of manage the attribute of the data structure of whole file system, the data structure of managing allocation of space, file and pointer, catalogue; The data of file system refer to the data of file; Take EXT2 as example, the content of superblock, piece group descriptor, inode bitmap, data block bitmap, inode table, catalogue all belongs to the metadata of file system, and the data of file system are the contents that leaves the file in data block district in;
Initial phase needs the metadata of buffer memory part, and as superblock, and the metadata of run duration buffer memory is need to replace time, adopts reading of coarsegrain; In service in file system, for searching of file or catalogue, when the metadata amount of needs access is during in byte number magnitude, adopt small grain size to read; In service in file system, the content of access file or catalogue, all adopts reading of coarsegrain;
In service in file system, while needing more new metadata and data for the deletion of file or catalogue, establishment, modification, rename etc., metadata on the medium of non-volatile small grain size addressing is adopted to writing back of small grain size, to data acquisition the writing back with coarsegrain on the medium of non-volatile coarsegrain addressing.
With a concrete example, the present invention is described in detail below:
1), by two media unified addressing, if the numbering of this two media is respectively A, B, wherein A is the non-volatile media of small grain size addressing, B is the non-volatile media of coarsegrain addressing, two media sequential addressing, establishing address is a～z, the address that wherein a～d is medium A, e～z is the address of medium B, and file system can record the boundary of this two classes physical address.Fig. 2 is a typical example that mixes storage, wherein the medium of non-volatile small grain size addressing is PCM, and the medium of non-volatile coarsegrain addressing is solid-state disk, supposes that the size that PCM can provide is 4K, the size that solid-state disk can provide is 4M, and address assignment is as Fig. 2;
2), in document layout system, the metadata of the media storage file system of non-volatile small grain size addressing, the data of the media storage file of non-volatile coarsegrain addressing.File system records the boundary of two media, and boundary both sides are metadata and the file content of storing documents system respectively.1) in, the metadata of address a～d memory file system, the real content of address e～z storage file.Take EXT2 as example, suppose to only have two groups, Fig. 3 sees in physical layout.On the medium of non-volatile small grain size read-write, deposit the metadata of EXT2 file system, comprise the metadata of piece group 0 and piece group 1, the metadata of each group comprises the data block of superblock, piece group descriptor, data block bitmap, inode bitmap, inode table and storage directory content; The data of storing documents system on the medium of non-volatile coarsegrain addressing, the i.e. content of storing documents.Need explanation a bit, scheme provided by the invention is not limited to the file system of EXT2 and series, adopts this scheme need to add to existing file system data structure and the corresponding read-write mechanism of part;
3), have reading of two kinds of granularities in file system, concrete size is relevant with medium.For the medium of coarsegrain addressing, all adopt reading of coarsegrain; For the medium of small grain size addressing, need cache metadata at initial phase, and the metadata of run duration buffer memory is need to replace time, adopts reading of coarsegrain, in search procedure, when the data of access catalogue, adopt reading of coarsegrain; In service in file system, when the metadata amount of needs access is during in byte number magnitude, adopt small grain size to read, as Fig. 4.Such as in Fig. 2, if adopt the layout of Fig. 3, in solid-state disk buffer memory the content of file, in the time of the content of access file, all adopt reading of coarsegrain, such as reading by page.For the metadata of buffer memory in PCM, in the time of initialization, need to import the information of superblock, part group descriptor and part bitmap, what now adopt is that coarsegrain is read, in operational process, as need are replaced these data, when searching, the piece group descriptor at place, not in internal memory, can be called in this piece group descriptor and replace the information of superseded piece group descriptor, and reading is now also that the coarsegrain adopting is read, the content that can travel through catalogue in search procedure is also to adopt reading of coarsegrain; In operational process, if only need the content of reading section, in search procedure, only need to know specific inode information, now can adopt reading of small grain size;
4), have writing of two kinds of granularities in file system, concrete size is relevant with medium.For the medium of coarsegrain addressing, all adopt writing of coarsegrain; For the medium of small grain size addressing, in service in file system, while needing more new metadata for the deletion of file or catalogue, establishment, modification, rename etc., metadata is adopted to writing back of small grain size, as Fig. 5.Such as in Fig. 2, if adopt the layout of Fig. 3, in solid-state disk buffer memory the content of file, in the time of the content of updating file, all adopt writing of coarsegrain, such as reading by page.For the metadata of buffer memory in PCM, in file system operational process, revise and upgrade these data as need, during such as a file of establishment, the operation that relates to metadata comprises No. inode, distribution, distribute data piece, writes inode, adds directory entry, and these all need to complete on PCM, now adopt granule to write, write such as adopting by byte, place, the bit position byte of upgrading in inode bitmap and data block bitmap, the inode content of renewal and the directory entry of interpolation are write back to PCM;
Those skilled in the art will readily understand; the foregoing is only preferred embodiment of the present invention; not in order to limit the present invention, all any modifications of doing within the spirit and principles in the present invention, be equal to and replace and improvement etc., within all should being included in protection scope of the present invention.
Claims (10)
1. the mixed file system based on different storage mediums, is characterized in that, described storage medium comprises the medium of non-volatile small grain size addressing and the medium of non-volatile coarsegrain addressing, and two media is unified sequential addressing; Wherein, the medium of described non-volatile small grain size addressing is for the metadata of storing documents system, and the medium of described non-volatile coarsegrain addressing is for the data of storing documents system; Described file system is called varigrained reading or writing in different situations to different address realms.
2. the system as claimed in claim 1, is characterized in that, described two media is unified sequential addressing and is specially: the medium of described non-volatile small grain size addressing is addressed in LAP Low Address Part, and the medium of described non-volatile coarsegrain addressing is addressed in high address section.
3. the system as claimed in claim 1, it is characterized in that, the metadata of described file system refers to the data structure in managed storage space, metadata with file in file system, and directory tree structure, specifically comprise attribute and the content of the attribute of data structure, file of data structure, the management allocation of space of managing whole file system and pointer, catalogue.
4. the system as claimed in claim 1, is characterized in that, the data of described file system refer to the content of the file that leaves data block district in.
5. the system as described in claim 1 to 4 any one, is characterized in that, described file system is called varigrained reading or writing in different situations to different address realms and comprised:
In the time that the file system initialization stage needs buffer memory part metadata, and the metadata of run duration buffer memory is need to replace time, adopts reading of coarsegrain; In service in file system, the content of access file or catalogue, adopts reading of coarsegrain.
6. the system as described in claim 1 to 4 any one, is characterized in that, described file system is called varigrained reading or writing in different situations to different address realms and comprised:
In service in file system, for searching of file or catalogue, when the metadata amount of needs access is during in byte number magnitude, adopt small grain size to read.
7. the system as described in claim 1 to 4 any one, is characterized in that, described file system is called varigrained reading or writing in different situations to different address realms and comprised:
In service in file system, when needs are more when new metadata, the metadata on the medium of non-volatile small grain size addressing is adopted to writing back of small grain size, wherein need the more situation of new metadata to comprise the deletion to file or catalogue, or create, or revise, or rename.
8. the system as described in claim 1 to 4 any one, is characterized in that, described file system is called varigrained reading or writing in different situations to different address realms and comprised:
In service in file system, in the time of needs batch updating data, with the writing back of coarsegrain, wherein need the more situation of new data to comprise the deletion to file to the data acquisition on the medium of non-volatile coarsegrain addressing, or create, or revise.
9. the system as claimed in claim 1, is characterized in that, the medium of described non-volatile small grain size addressing is phase transition storage, and the medium of described non-volatile coarsegrain addressing is solid-state disk.
10. the system as claimed in claim 1, is characterized in that, if described file system is EXT2 form, the metadata of described file system comprises the content of superblock, piece group descriptor, inode bitmap, data block bitmap, inode table and catalogue.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410096969.3A CN103838853B (en) | 2014-03-17 | 2014-03-17 | Mixed file system based on different storage media |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410096969.3A CN103838853B (en) | 2014-03-17 | 2014-03-17 | Mixed file system based on different storage media |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103838853A true CN103838853A (en) | 2014-06-04 |
| CN103838853B CN103838853B (en) | 2017-05-10 |
Family
ID=50802350
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410096969.3A Active CN103838853B (en) | 2014-03-17 | 2014-03-17 | Mixed file system based on different storage media |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103838853B (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104123264A (en) * | 2014-08-01 | 2014-10-29 | 浪潮(北京)电子信息产业有限公司 | Cache management method and device based on heterogeneous integrated framework |
| CN104536903A (en) * | 2014-12-25 | 2015-04-22 | 华中科技大学 | Mixed storage method and system for conducting classified storage according to data attributes |
| CN104537050A (en) * | 2014-12-25 | 2015-04-22 | 华中科技大学 | Method for batch and rapid establishment of metadata and data of file system |
| WO2016041156A1 (en) * | 2014-09-17 | 2016-03-24 | 华为技术有限公司 | Method and apparatus for scheduling cpu |
| CN106155932A (en) * | 2016-07-04 | 2016-11-23 | 深圳市硅格半导体股份有限公司 | Data cache method based on Flash and device |
| CN106599236A (en) * | 2016-12-20 | 2017-04-26 | 北海市云盛科技有限公司 | Metadata storage method and apparatus for file system |
| CN107590285A (en) * | 2017-09-30 | 2018-01-16 | 郑州云海信息技术有限公司 | A kind of method of heterogeneous system data consistency |
| CN107766002A (en) * | 2017-10-20 | 2018-03-06 | 华中科技大学 | A kind of virtual mixed file system based on mixing storage device |
| CN107766374A (en) * | 2016-08-19 | 2018-03-06 | 上海凯翔信息科技有限公司 | The optimization method and system that a kind of mass small documents storage is read |
| CN108595591A (en) * | 2018-04-19 | 2018-09-28 | 国云科技股份有限公司 | A kind of method of mixed file storage |
| CN110727403A (en) * | 2019-09-12 | 2020-01-24 | 华为技术有限公司 | Metadata management method and device |
| CN111190550A (en) * | 2019-12-31 | 2020-05-22 | 深圳市安云信息科技有限公司 | Metadata acceleration method and device and storage equipment |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006092124A (en) * | 2004-09-22 | 2006-04-06 | Fujitsu Ltd | Storage device, storage control method and storage control program |
| CN102075581B (en) * | 2011-01-25 | 2013-06-12 | 中国科学院计算技术研究所 | Data transmission method and device oriented to distributed file system |
| CN102681884A (en) * | 2011-03-17 | 2012-09-19 | 新奥特（北京）视频技术有限公司 | Method for solving medium asset metadata read-write access conflicts |
| CN102364474B (en) * | 2011-11-17 | 2014-08-20 | 中国科学院计算技术研究所 | Metadata storage system for cluster file system and metadata management method |
-
2014
- 2014-03-17 CN CN201410096969.3A patent/CN103838853B/en active Active
Cited By (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104123264A (en) * | 2014-08-01 | 2014-10-29 | 浪潮(北京)电子信息产业有限公司 | Cache management method and device based on heterogeneous integrated framework |
| CN105612505B (en) * | 2014-09-17 | 2018-12-14 | 华为技术有限公司 | The method and apparatus of CPU scheduling |
| WO2016041156A1 (en) * | 2014-09-17 | 2016-03-24 | 华为技术有限公司 | Method and apparatus for scheduling cpu |
| CN105612505A (en) * | 2014-09-17 | 2016-05-25 | 华为技术有限公司 | Method and apparatus for scheduling CPU |
| CN104536903A (en) * | 2014-12-25 | 2015-04-22 | 华中科技大学 | Mixed storage method and system for conducting classified storage according to data attributes |
| CN104537050A (en) * | 2014-12-25 | 2015-04-22 | 华中科技大学 | Method for batch and rapid establishment of metadata and data of file system |
| CN104537050B (en) * | 2014-12-25 | 2017-12-15 | 华中科技大学 | A kind of method of the quick establishment file system metadata of batch and data |
| CN104536903B (en) * | 2014-12-25 | 2018-02-23 | 华中科技大学 | A kind of mixing storage method and system stored classifiedly by data attribute |
| CN106155932A (en) * | 2016-07-04 | 2016-11-23 | 深圳市硅格半导体股份有限公司 | Data cache method based on Flash and device |
| CN107766374B (en) * | 2016-08-19 | 2021-05-25 | 上海凯翔信息科技有限公司 | Optimization method and system for storage and reading of massive small files |
| CN107766374A (en) * | 2016-08-19 | 2018-03-06 | 上海凯翔信息科技有限公司 | The optimization method and system that a kind of mass small documents storage is read |
| CN106599236A (en) * | 2016-12-20 | 2017-04-26 | 北海市云盛科技有限公司 | Metadata storage method and apparatus for file system |
| CN107590285A (en) * | 2017-09-30 | 2018-01-16 | 郑州云海信息技术有限公司 | A kind of method of heterogeneous system data consistency |
| CN107766002A (en) * | 2017-10-20 | 2018-03-06 | 华中科技大学 | A kind of virtual mixed file system based on mixing storage device |
| CN108595591A (en) * | 2018-04-19 | 2018-09-28 | 国云科技股份有限公司 | A kind of method of mixed file storage |
| CN110727403A (en) * | 2019-09-12 | 2020-01-24 | 华为技术有限公司 | Metadata management method and device |
| CN111190550A (en) * | 2019-12-31 | 2020-05-22 | 深圳市安云信息科技有限公司 | Metadata acceleration method and device and storage equipment |
| CN111190550B (en) * | 2019-12-31 | 2024-03-29 | 深圳市安云信息科技有限公司 | Metadata acceleration method and device and storage equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103838853B (en) | 2017-05-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103838853A (en) | Mixed file system based on different storage media | |
| US9298384B2 (en) | Method and device for storing data in a flash memory using address mapping for supporting various block sizes | |
| US10282286B2 (en) | Address mapping using a data unit type that is variable | |
| CN103902669B (en) | A kind of separate type file system based on different storage mediums | |
| CN101840308B (en) | Hierarchical memory system and logical volume management method thereof | |
| CN107168657B (en) | Virtual disk hierarchical cache design method based on distributed block storage | |
| CN106708427A (en) | Storage method suitable for key value pair data | |
| US8650379B2 (en) | Data processing method for nonvolatile memory system | |
| CN102779096B (en) | Page, block and face-based three-dimensional flash memory address mapping method | |
| EP2735978A1 (en) | Storage system and management method used for metadata of cluster file system | |
| US20180349285A1 (en) | Managing Multiple Namespaces in a Non-Volatile Memory (NVM) | |
| CN110347336A (en) | A kind of key assignments storage system based on NVM with SSD mixing storage organization | |
| CN103186350A (en) | Hybrid storage system and hot spot data block migration method | |
| WO2012126317A1 (en) | File storage method and device thereof | |
| CN105138286A (en) | Method for mixed utilization of SSD and SMR hard disks in disk file system | |
| CN103514249A (en) | Method and system for automatic data reduction and storage device | |
| CN102915340A (en) | Expanded B+ tree-based object file system | |
| CN111522507B (en) | Low-delay file system address space management method, system and medium | |
| CN105045850B (en) | Junk data recovery method in cloud storage log file system | |
| CN109240944B (en) | Data reading and writing method based on variable-length cache line | |
| CN108052541B (en) | File system implementation and access method based on multi-level page table directory structure and terminal | |
| CN102201005A (en) | Method for distributing streaming media enhanced file of file system | |
| CN102902709A (en) | Space allocation fixing file memory system and implementation method | |
| CN103049224A (en) | Method, device and system for importing data into physical tape | |
| CN102520885B (en) | Data management system for hybrid hard disk |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |