CN1632779A - A data storage method and apparatus - Google Patents
A data storage method and apparatus Download PDFInfo
- Publication number
- CN1632779A CN1632779A CN 200310121069 CN200310121069A CN1632779A CN 1632779 A CN1632779 A CN 1632779A CN 200310121069 CN200310121069 CN 200310121069 CN 200310121069 A CN200310121069 A CN 200310121069A CN 1632779 A CN1632779 A CN 1632779A
- Authority
- CN
- China
- Prior art keywords
- data
- data recording
- storage
- recording
- record
- 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
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
This invention discloses a data memory method, which comprises the following steps: to allocate one ID for each data record; to get the index memory address of the data record according to the ID mapping; to store the data record into one divided files; to store the first address in the said index memory area. When the outer application system visits the data record, it runs the data record relative ID to realize the data record memory, visiting and deleting.
Description
Technical field
The present invention relates to field of data storage, relate in particular to a kind of high-level efficiency date storage method and device.
Background technology
In current cybertimes, the Network service system is more and more common, and along with popularizing that broadband network, 3G technology are used, the storage of content of multimedia is more and more important to these systems.As shown in Figure 1, the Network service system is when the user provides multimedia service, and this system is storage multimedia earlier, wait to receive user A send obtain the multimedia request after, inquire about specific multimedia, provide Query Result so that multimedia service to be provided to user A then.In the multimedia service shown in Figure 2 for another example, also relate to multimedia storage.The application of multimedia storage also has a lot, does not enumerate one by one here.In a word, network service system all needed the multimedia recording of quick storage, inquiry, deletion correspondence before providing service for the client.
In Internet, incomplete epoch of radio network technique, the network bandwidth is very restricted, the literal that the medium of client access, forwarding are brief often, network application system often use the general commercial database can realize the storage of medium, fast access.
Along with network technology is more and more perfect, use more and more universal, the medium that transmitted no longer only are confined to brief word content, but the greatly abundant content of multimedia of the content and the form of expression, comprise sound, image, film, commercial file etc., and need the multimedia recording huge amount of storage, wall scroll record size is indefinite, does not often wait to hundreds of K from several K.It is unable to do what one wishes that the general commercial database is used for storing these multi-medium datas, and inefficiency is inapplicable for those systems that efficient is had relatively high expectations.The system that has transfers directly to seek help from operating system, directly adopts operating system file to come medium, and concrete scheme is: a file visits Multimedia Message according to filename later on corresponding to a Multimedia Message.This scheme is less demanding for those efficient, mini-system still can, but can't satisfy the requirement of large scale system because this scheme causes existing inevitably in the system heap file, operating system burden greatly increases, thereby causes the entire system performance low.And operating system increases, the delete file operation performance is not high, and can produce disk fragments inevitably, makes the performance of system go from bad to worse.In a word, for huge amount, the multimedia storage demand that wall scroll record size is little, this is not a feasible program.
The system that has then adopts database+big documents programme: with multimedia, storage and uniform is in one (or one group) big file, and the start address that concrete Multimedia Message is deposited leaves in the database.This scheme need increase the commercial data base system of a costliness, and cost is very high; And must toss about in bed repeatedly accessing database, multimedia file and just can have access to content of multimedia, thereby efficient is not fully up to expectations.And content of multimedia storage organization in big file is equally complicated, and the use because of database does not reduce.
Summary of the invention
At the above-mentioned defective of prior art, the object of the present invention is to provide quick storage, fast access, the deletion fast of a kind of date storage method and device, and cost is low to realize data.
The invention provides a kind of method of data storage, this method is: for data recording to be stored is distributed a data record identification (ID); Described data recording is kept in the file that is divided into bunch; The address of storing first bunch of described data recording is stored in the index stores district, and sets up the mapping relations one by one of this address and data recording ID.
Wherein:
Set up the idle storehouse of data recording ID, when a certain data recording of deletion, the ID of this record is positioned in the idle storehouse of data recording ID, and the maximal value of preserving the ID that has dispensed.
During for data recording distribute data Record ID to be stored, at first in the idle storehouse of judgment data Record ID whether idle ID is arranged,, then preferentially use this free time ID, as the maximal value of the ID that does not have idle ID then to continue to have dispensed generates new ID if having.
For the file of store data record bunch set up free cluster pointer storehouse, when a certain data recording of deletion, bunch pointer of the correspondence of this record is positioned in the free cluster pointer storehouse, and preserves the maximal value of bunch pointer that has dispensed.
When storing described data recording hereof, at first in the judgment data record free cluster storehouse whether the free cluster pointer is arranged,, then preferentially use this free cluster if having; Otherwise, bunch the using of having dispensed of continuing new bunch.
When visit data write down, the ID by this data recording visited described record.
Preserve in all bunches of a data record, each bunch preserved next bunch address.
The invention provides a kind of data storage device, described device comprises Service Processing Module, storage control module and physical storage of data district,
Described Service Processing Module is used to accept the data recording operation request, and finishes the mutual conversion of the application specific data structure of described data recording to binary data blocks, and the binary data record block after the conversion will be stored by storage control module;
Described storage control module is used to described data recording distribute data Record ID, and sets up the relation that this data recording ID and data are recorded in physical storage of data district memory address;
Described physical storage of data district is used to store all data recording.
Described Service Processing Module comprises the business processing stubs, stringization and playing a reversed role that described stubs is used to provide a class and finishes data recording.
Described physical storage of data district comprises index stores district and data storage area, is respectively applied for index and data recording with document form storage data recording; The data file that the data storage area forms be split into one by one bunch, each bunch preserved next bunch address; Described data storage area also comprises the free cluster storehouse, is used for when data recording is deleted, and preserves the pointer of corresponding bunch of this record; Described index stores district also comprises data recording ID storehouse, is used for when data recording is deleted, and preserves the corresponding ID of this record.
Described storage control module further comprises storage manager and at least one packet memory manager, and described storage manager is used for giving packeting memory with the storage administration work allocation of reality, the corresponding worker thread of each packet manager.
Described packeting memory further comprises index stores manager and data storage manager,
Described index stores manager is used to manage idle data Record ID storehouse and index stores district, when increasing data recording newly, at first is the idle ID of this record application from idle ID storehouse, were it not for the ID number new ID of generation that idle ID then continues and distributed;
Described data storage manager is used to manage free cluster pointer storehouse and data storage area, when newly-increased data recording, at first applies for free cluster for this record from free cluster pointer storehouse, were it not for free cluster then with this recording storage in new bunch.
Described data recording ID is obtained by mapping transformation one by one by the corresponding file side-play amount of this recording indexes memory address.
Described data recording is a multimedia recording.
The index stores manager further comprises data recording ID manager, is used to manage idle data Record ID storehouse.
Described a kind of data storage device links to each other with application system, accepts the storage request of application system, according to request store accordingly, visit and deletion action, application system is by data recording ID visit data record.
The present invention is that data recording distributes ID number, and this ID is obtained by mapping transformation one by one by this index data record memory block corresponding file side-play amount.When application system visited by this ID, the convertible corresponding stored regional address that obtains was avoided the complicated query hunting action fully, reaches maximizing efficiency.And data recording storage is to be finished by the packet memory manager, making full use of the I/O of system bandwidth, and does not need expensive commercial data base system support, and cost is low.
Description of drawings
Fig. 1 is multimedia application sample one process flow diagram;
Fig. 2 is multimedia application sample two process flow diagrams;
Fig. 3 is the data storage device architectural schematic;
Fig. 4 is the storage control module architectural schematic;
Fig. 5 is the data storage area structural representation;
Fig. 6 is idle ID stack algorithm synoptic diagram;
Fig. 7 is a storage means process flow diagram among the present invention program.
Embodiment
For making the purpose, technical solutions and advantages of the present invention clearer, in data recording, select multimedia recording as embodiment, the present invention is described in more detail with accompanying drawing in conjunction with the embodiments.
As shown in Figure 3, data storage device provided by the invention links to each other with application system, accepts the storage request of application system, according to request store accordingly, visit and deletion action.
This data storage device comprises Service Processing Module, storage control module, physical storage of data district.Wherein Service Processing Module comprises the business processing stubs again, and it provides a class to application system, and this stake provides stringization (marshaling), playing a reversed role (demarshaling) function, finishes the mutual conversion of application specific data structure to binary data blocks.It is convenient to storage control module and realizes unified storage administration to the diversity of the data structure of " storage control module " shielding application system.
As shown in Figure 4, storage control module is made up of storage manager and packet memory manager.Storage manager plays storage management agents, and it goes to finish to the packet memory manager storage administration working delegation of reality.It provides unified virtual multimedia memory device by the business processing stubs to application system.It is that every multimedia recording distributes a multimedia recording sign (ID), and finishes the binding of multimedia recording ID to data block.This ID can be designed as the numerical value of a fixed length, and is as shown in the table, and it amounts to 12 figure places, and 11～12 is " packet number ", 1～10th, and span suitably be dwindled or be enlarged to unified numbering can during actual certainly enforcement, and all see the goal systems scale.When application system visit multimedia recording, conduct interviews according to ID number of this multimedia recording correspondence.
| 12 | ????11 | ????10 | ????9 | ????8 | ????7 | ????6 | ????5 | ????4 | ????3 | ????2 | |
| Segment number | Unified numbering | ||||||||||
Packet memory manager: specifically finish the storage administration task.With the main target of storage administration grouping be: content of multimedia can be stored in respectively on the different files by grouping, and then can be mapped on the different disks, so, can improve the I/O of system efficient.Each packet memory manager has oneself a worker thread, can also realize parallel storage action so easily.Certainly, also can be on identical file with all multimedia data storage, in this case, storage control module does not need the packet memory manager, finishes concrete storage administration task by storage manager.
Physical storage of data district: be divided into index stores district and multi-medium data memory block, the big document storage mode of unified employing.Wherein the index stores district stores the index of each multimedia recording, also comprises multimedia recording ID storehouse, is used for when multimedia recording is deleted, and preserves the corresponding ID of this record.When multimedia recording of application system request storage, the storage administration floor is that the ID that distributes of every multimedia recording is obtained by mapping transformation one by one by index stores district corresponding file side-play amount.Near and, set up the ID of each multimedia recording and this multimedia recording one-to-one relationship by the index stores district at the index of multi-medium data memory block.The multi-medium data memory block adopts big file to add scheme by bunch distribution, be about to the big file division of data become one by one bunch, each bunch preserved next bunch address, implementation is analogous to chained list.In the index area, preserve the address of first bunch of multimedia recording correspondence.So, set up the ID of each multimedia recording and this multimedia recording one-to-one relationship at the index of multi-medium data memory block by the index stores district.The address of first bunch by the multimedia recording correspondence of preserving in the index area and then have access to the associated multimedia record data again.Described data storage area also comprises free cluster pointer storehouse, is used for when multimedia recording is deleted, and preserves the corresponding bunch pointer of this record.The physical storage of data plot structure as shown in Figure 5.
In said system, storage control module is the core of whole multimedia memory storage, each packet memory manager inside comprises index stores manager and data storage manager again, and comprises multimedia recording ID manager and free cluster pointer management device in index stores manager and the data storage manager accordingly.The index stores manager is used to manage index datastore district and idle multimedia recording ID storehouse.In the time will deleting certain multimedia recording, the ID of this record is placed in the idle multimedia recording ID storehouse, when newly-increased record, at first application multimedia recording ID from the ID storehouse of free time were it not for the new multimedia recording ID of idle ID regeneration.The multimedia data storage management device is used to manage multimedia memory block and idle bunch pointer, when certain record of deletion, should write down pairing bunch of pointer is placed in the free cluster pointer storehouse, when increasing multimedia recording newly, at first from bunch pointer storehouse of free time, apply for resource, when were it not for free cluster, regeneration new bunch.
Wherein, multimedia recording ID manager is the key that the storage administration layer is realized, must at first apply for obtaining ID during Server, Portal visit multimedia recording memory storage, must reclaim ID after the storage data discharge.It realizes that principle is the inner idle ID storehouse of internal memory of preserving of ID manager, and corresponding backup file storehouse is arranged in the file, preserves an ID maximal value that has dispensed simultaneously, the ID that preferential use has been reclaimed.And internal memory storehouse and respective file reinforcement storehouse ought carry out with batch read-write form alternately, as shown in Figure 6.
Outer application system is in visit during multimedia recording, visits with multimedia recording ID and avoids the complicated query hunting action.
The invention provides a kind of date storage method, as shown in Figure 7, its core is: for every data recording is distributed an ID, can obtain the index stores regional address of this multidata record according to this ID mapping transformation, this data recording is kept at one is divided in many bunches the file, the address of first bunch of this multimedia recording of storage is stored in the described index stores district.This ID is the numerical value of a fixed length, and span can suitably be dwindled or enlarge to unified numbering during actual certainly enforcement.When deletion during a certain multimedia recording, the ID of this record is positioned in the idle storehouse of corresponding multimedia recording ID.
Set up the idle storehouse of multimedia recording ID earlier, when a certain multimedia recording of deletion, the ID of this record is positioned in the idle storehouse of corresponding multimedia recording ID, and the maximal value of preserving the ID that has dispensed.When newly-increased multimedia recording, judge at first in the idle storehouse of multimedia recording ID whether idle ID is arranged, if having, then preferentially use this free time ID, were it not for then the continue maximal value of the ID that dispensed of idle ID and generate new ID.
Bunch set up free cluster pointer storehouse for what deposit multimedia file, when a certain multimedia recording of deletion, bunch pointer of the correspondence of this record be pressed in the free cluster storehouse, and preserve dispensed bunch maximal value.When newly-increased multimedia recording, whether in free cluster storehouse have free cluster, if having, then preferentially use this free cluster if at first judging, were it not for free cluster then continue bunch using of having dispensed new bunch.
In sum, when needs are stored a multimedia recording,, divide the timing priority allocation idle ID for multimedia recording distributes a multimedia recording ID, as do not have idle ID then to apply for new ID, according to the convertible index stores regional address that obtains this multimedia recording of this ID mapping.With multimedia recording storage hereof, during storage, preferentially be stored in free time in this document bunch, if there is not free cluster, then be stored in new bunch, the address of this bunch is stored in the index stores regional address of this multimedia recording.When outer application system visit multimedia recording, the ID by this multimedia recording correspondence carries out, and then has realized quick storage, visit and the deletion of this multimedia recording.
The above only is the present invention program's preferred embodiment, not in order to limit protection scope of the present invention.
Claims (18)
1, a kind of method of data storage is characterized in that the method comprising the steps of:
For data recording to be stored is distributed a data record identification (ID);
Described data recording is kept in the file that is divided into bunch;
The address of storing first bunch of described data recording is stored in the index stores district, and sets up the mapping relations one by one of this address and data recording ID.
2, the method for claim 1 is characterized in that, sets up the idle storehouse of data recording ID, when a certain data recording of deletion, the ID of this record is positioned in the idle storehouse of data recording ID, and the maximal value of preserving the ID that has dispensed.
3, method as claimed in claim 2, it is characterized in that: during for data recording distribute data Record ID to be stored, at first in the idle storehouse of judgment data Record ID whether idle ID is arranged, if have, then preferentially use this free time ID, as the maximal value of the ID that does not have idle ID then to continue to have dispensed generates new ID.
4, the method for claim 1, it is characterized in that: bunch set up free cluster pointer storehouse for the file of store data record, when a certain data recording of deletion, bunch pointer of the correspondence of this record is positioned in the free cluster pointer storehouse, and preserves the maximal value of bunch pointer that has dispensed.
5, method as claimed in claim 4 is characterized in that: when storing described data recording hereof, at first in the judgment data record free cluster storehouse whether the free cluster pointer is arranged, if having, then preferentially use this free cluster; Otherwise, bunch the using of having dispensed of continuing new bunch.
6, the method for claim 1 is characterized in that: when visit data write down, the ID by this data recording visited described record.
7, the method for claim 1 is characterized in that: preserve in all bunches of a data record, each bunch preserved next bunch address.
8, as the arbitrary described method of claim 1 to 7, it is characterized in that: described data recording ID is the numerical value of certain-length.
9, method as claimed in claim 8 is characterized in that: described data recording is meant multimedia recording.
10, a kind of data storage device is characterized in that: described device comprises Service Processing Module, storage control module and physical storage of data district;
Described Service Processing Module is used to accept the data recording operation request, and finishes the mutual conversion of the application specific data structure of described data recording to binary data blocks, and the binary data record block after the conversion will be stored by storage control module;
Described storage control module is used to described data recording distribute data Record ID, and sets up the relation that this data recording ID and data are recorded in physical storage of data district memory address;
Described physical storage of data district is used to store all data recording.
11, device as claimed in claim 10 is characterized in that: described Service Processing Module comprises the business processing stubs, stringization and playing a reversed role that described stubs is used to provide a class and finishes data recording.
12, device as claimed in claim 10 is characterized in that: described physical storage of data district comprises index stores district and data storage area, is respectively applied for index and data recording with document form storage data recording; The data file that the data storage area forms be split into one by one bunch, each bunch preserved next bunch address; Described data storage area also comprises the free cluster storehouse, is used for when data recording is deleted, and preserves the pointer of corresponding bunch of this record; Described index stores district also comprises data recording ID storehouse, is used for when data recording is deleted, and preserves the corresponding ID of this record.
13, device as claimed in claim 10, it is characterized in that: described storage control module further comprises storage manager and at least one packet memory manager, described storage manager is used for giving packeting memory with the storage administration work allocation of reality, the corresponding worker thread of each packet manager.
14, device as claimed in claim 13 is characterized in that: described packeting memory further comprises index stores manager and data storage manager;
Described index stores manager is used to manage idle data Record ID storehouse and index stores district, when increasing data recording newly, at first is the idle ID of this record application from idle ID storehouse, were it not for the ID number new ID of generation that idle ID then continues and distributed;
Described data storage manager is used to manage free cluster pointer storehouse and data storage area, when newly-increased data recording, at first applies for free cluster for this record from free cluster pointer storehouse, were it not for free cluster then with this recording storage in new bunch.
15, as each described device in the claim 10 to 14, it is characterized in that: described data recording ID is obtained by mapping transformation one by one by the corresponding file side-play amount of this recording indexes memory address.
16, as each described device in the claim 10 to 14, it is characterized in that: described data recording is a multimedia recording.
17, as each described device in the claim 10 to 14, it is characterized in that: described a kind of data storage device links to each other with application system, accept the storage request of application system, according to request store accordingly, visit and deletion action, application system is by data recording ID visit data record.
18, device as claimed in claim 14 is characterized in that: the index stores manager further comprises data recording ID manager, is used to manage idle Record ID storehouse.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2003101210691A CN100342374C (en) | 2003-12-24 | 2003-12-24 | A data storage method and apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2003101210691A CN100342374C (en) | 2003-12-24 | 2003-12-24 | A data storage method and apparatus |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1632779A true CN1632779A (en) | 2005-06-29 |
| CN100342374C CN100342374C (en) | 2007-10-10 |
Family
ID=34844034
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2003101210691A Expired - Fee Related CN100342374C (en) | 2003-12-24 | 2003-12-24 | A data storage method and apparatus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100342374C (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009076854A1 (en) * | 2007-12-03 | 2009-06-25 | Tencent Technology (Shenzhen) Company Limited | Data cache system and method for realizing high capacity cache |
| WO2010148549A1 (en) * | 2009-06-22 | 2010-12-29 | Sheng Yongxiang | System for identifying a user and its maintaining method |
| CN101645039B (en) * | 2009-06-02 | 2011-06-22 | 中国科学院声学研究所 | Method for storing and reading data based on Peterson graph |
| CN101295169B (en) * | 2007-01-29 | 2012-01-04 | 洛克威尔自动控制技术股份有限公司 | Method for indirect access to controller data using name stored in string tag |
| CN102377586A (en) * | 2010-08-16 | 2012-03-14 | 研祥智能科技股份有限公司 | Network bypass device and method for processing network bypass |
| CN101105800B (en) * | 2007-07-04 | 2012-12-05 | 深圳市中兴移动通信有限公司 | High capacity data record storage method for embedded system |
| CN102033875B (en) * | 2009-09-24 | 2012-12-12 | 北大方正集团有限公司 | File warehousing method and device |
| CN101763310B (en) * | 2008-12-23 | 2013-10-23 | 康佳集团股份有限公司 | Data storage control method |
| CN104516988A (en) * | 2015-01-21 | 2015-04-15 | 天津书生云科技有限公司 | File writing method and device |
| CN107710151A (en) * | 2015-06-24 | 2018-02-16 | 英特尔公司 | The technology that shadow storehouse for binary file converting system manipulates |
| CN117453632A (en) * | 2023-12-25 | 2024-01-26 | 杭州计算机外部设备研究所(中国电子科技集团公司第五十二研究所) | Data storage method and device |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH05313980A (en) * | 1992-05-07 | 1993-11-26 | Olympus Optical Co Ltd | Information recording method |
| DE69431979T2 (en) * | 1993-06-14 | 2003-10-30 | Sony Corp | ADMINISTRATIVE PROCEDURE FOR A RECORDING MEDIUM |
-
2003
- 2003-12-24 CN CNB2003101210691A patent/CN100342374C/en not_active Expired - Fee Related
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101295169B (en) * | 2007-01-29 | 2012-01-04 | 洛克威尔自动控制技术股份有限公司 | Method for indirect access to controller data using name stored in string tag |
| CN101105800B (en) * | 2007-07-04 | 2012-12-05 | 深圳市中兴移动通信有限公司 | High capacity data record storage method for embedded system |
| WO2009076854A1 (en) * | 2007-12-03 | 2009-06-25 | Tencent Technology (Shenzhen) Company Limited | Data cache system and method for realizing high capacity cache |
| CN101763310B (en) * | 2008-12-23 | 2013-10-23 | 康佳集团股份有限公司 | Data storage control method |
| CN101645039B (en) * | 2009-06-02 | 2011-06-22 | 中国科学院声学研究所 | Method for storing and reading data based on Peterson graph |
| WO2010148549A1 (en) * | 2009-06-22 | 2010-12-29 | Sheng Yongxiang | System for identifying a user and its maintaining method |
| CN102576359A (en) * | 2009-06-22 | 2012-07-11 | 深圳市永盛世纪科技有限公司 | System for identifying a user and its maintaining method |
| CN102033875B (en) * | 2009-09-24 | 2012-12-12 | 北大方正集团有限公司 | File warehousing method and device |
| CN102377586A (en) * | 2010-08-16 | 2012-03-14 | 研祥智能科技股份有限公司 | Network bypass device and method for processing network bypass |
| CN104516988A (en) * | 2015-01-21 | 2015-04-15 | 天津书生云科技有限公司 | File writing method and device |
| CN104516988B (en) * | 2015-01-21 | 2018-09-28 | 天津书生云科技有限公司 | A kind of file wiring method and device |
| CN107710151A (en) * | 2015-06-24 | 2018-02-16 | 英特尔公司 | The technology that shadow storehouse for binary file converting system manipulates |
| CN117453632A (en) * | 2023-12-25 | 2024-01-26 | 杭州计算机外部设备研究所(中国电子科技集团公司第五十二研究所) | Data storage method and device |
| CN117453632B (en) * | 2023-12-25 | 2024-04-12 | 杭州计算机外部设备研究所(中国电子科技集团公司第五十二研究所) | Data storage method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100342374C (en) | 2007-10-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1292370C (en) | Method and apparatus for data processing | |
| CN101039278A (en) | Data management method and system | |
| US7058783B2 (en) | Method and mechanism for on-line data compression and in-place updates | |
| CN100342374C (en) | A data storage method and apparatus | |
| CN109547566B (en) | Multithreading uploading optimization method based on memory allocation | |
| CN106708427A (en) | Storage method suitable for key value pair data | |
| CN1726454A (en) | A distributed network attached storage system | |
| CN101076021A (en) | Network data storage system | |
| CN101027668A (en) | Distributed storage architecture based on block map caching and VFS stackable file system modules | |
| CN101079034A (en) | System and method for eliminating redundancy file of file storage system | |
| CN103412884B (en) | The management method of embedded database under a kind of isomery storage medium | |
| TW201025114A (en) | File system for storage device which uses different cluster sizes | |
| CN102411634B (en) | A kind of date storage method promoting embedded database real-time | |
| US10365849B2 (en) | Dual granularity dynamic mapping with packetized storage | |
| CN1828556A (en) | Method, system and apparatus for data memory | |
| WO2010125574A1 (en) | Description | |
| CN1869933A (en) | Computer processing system for implementing data update and data updating method | |
| CN1851671A (en) | Method for saving global varible internal memory space | |
| CN1614591A (en) | Method for organizing and accessing distributive catalogue of document system | |
| CN1255748C (en) | Metadata hierarchy management method and system of storage virtualization system | |
| CN101068341A (en) | Stream media dispatching system and medium file scheduling method thereof | |
| CN1823327A (en) | Data area managing method in information recording medium and information processor employing data area managing method | |
| CN1547137A (en) | Mass file management system and method based on database | |
| CN111694765A (en) | Mobile application feature-oriented multi-granularity space management method for nonvolatile memory file system | |
| CN1920796A (en) | Cache method and cache system for storing file's data in memory blocks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20071010 Termination date: 20171224 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |