CN100583832C - Data management method and system - Google Patents
Data management method and system Download PDFInfo
- Publication number
- CN100583832C CN100583832C CN200710090676A CN200710090676A CN100583832C CN 100583832 C CN100583832 C CN 100583832C CN 200710090676 A CN200710090676 A CN 200710090676A CN 200710090676 A CN200710090676 A CN 200710090676A CN 100583832 C CN100583832 C CN 100583832C
- Authority
- CN
- China
- Prior art keywords
- record
- data
- lock
- memory block
- type
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/217—Database tuning
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a data management method. According to the number of the data to choose the type of a storage block, the data is stored in the large files comprising plurality of the same type storage blocks, and the data search and positioning is well done by the record identification of the data and the relevance of the storage address information. The invention also provides with corresponding data management system. Because a suitable type of the storage block can be chosen according to the number of the data, it is helpful to efficiently use the disk space as much as possible; Because the large files is used for data storage, it reduces the created file number in the operation system; Because the positioning way of the record identification and the relevance of the storage address information is used, the positioning time is a constant grade with a stable performance and can not decrease as much because of the increase of the number of data.
Description
Technical field
The present invention relates to the data management technique field, be specifically related to data managing method and system, the present invention can be used in the communication system service end customer multi-media information data of big quantity is stored and operated.
Background technology
Along with popularizing of broadband network, the third generation (3G:The 3rd Generation) mechanics of communication, the information data quantity of transmitting in communication network is increasing, content is also more and more abundanter, no longer be confined to simple information such as note, text, but develop into content and all very big abundant multimedia information of the form of expression gradually, comprise audio frequency, image, video etc.One of characteristics of multi-medium data are that the magnitude range difference of each data is bigger, do not wait to millions of bytes (MB) from several kilobytes (KB).
In communication system, provide professional service end usually to need a large amount of user profile data are managed, for example the multimedia information data of transmission is temporarily stored, or to operations such as data execution reading and writing.The existing data management mode that adopts mainly contains following two kinds:
A kind of mode that is based on relational database: multimedia information data is stored in the operating system file of database outside with big object (LOB) or binary large object forms such as (BLOB), manages by the data of relational database to storage.Under this mode, the operating system file that carries out storage generally there is not specific structural requirement, relational database is stored the location of data by relational calculus, under the very big situation of data volume, the time loss that relational calculus causes often is exponential growth, causes satisfying the performance requirement of operation system to accessing multimedia data.In addition, relational database generally also needs to carry out some extra inspection work in order purpose such as to guarantee data integrity, and has reduced the efficient of data management.
A kind ofly be based on the mode that small documents is deposited: this mode is used for the management of Email (Email) more.The Qmail system that an exemplary that adopts this mode is extensive use under the Unix system.In this system, adopt the small documents mode to deposit Email, add the filename that recipient information constitutes storage file with outbox side, be used to carry out the Search and Orientation of file.Quantity of documents under the characteristic that is limited to present operating system, catalogue can not be too many, and can influence the locating speed of file when quantity of documents is too much, so this scheme is just not too suitable under the huge situation of the data bulk of needs storage.
Summary of the invention
The embodiment of the invention provides a kind of high performance data managing method, comprising: according to the size of data to be stored, determine the quantity and the type of required memory block, the type of described memory block is divided according to size; With described data deposit in provide the big file of definite type stores piece, described big file is made up of some memory blocks of the same type; Preserve the incidence relation of record identification and its storage address information of described data.
The embodiment of the invention also provides corresponding data management system, comprising: the piece memory cell, be used to provide various dissimilar big files, and every type big file is made up of some memory blocks of the same type, and the type of described memory block is divided according to size; The storage policy unit is used for the size according to data to be stored, determines the quantity and the type of required memory block; The storage performance element, be used for memory block quantity and the type definite according to described storage policy unit, with data deposit in providing of described memory cell the big file of definite type stores piece, and preserve the incidence relation of record identification and its storage address information of described data.
The embodiment of the invention adopts the type of selecting memory block according to size of data, with storage in the big file of forming by some memory blocks of the same type, by the record identification of data and the related method of carrying out the Search and Orientation of data of its storage address information; Owing to can select suitable memory block type according to size of data, can try one's best and effectively utilize disk space and improve readwrite performance; Owing to adopt big file to carry out storage, reduced the quantity of documents of in operating system, creating; Because adopt the record identification locate mode related with storage address information, be the constant level positioning time, stable performance can not be subjected to influence that data volume increases and decline significantly.
Description of drawings
Fig. 1 is the embodiment of the invention one a data managing method schematic flow sheet;
Fig. 2 is a big file structure schematic diagram in the embodiment of the invention one method;
Fig. 3 is a data manipulation schematic flow sheet in the embodiment of the invention two data managing methods;
Fig. 4 is the embodiment of the invention three data management system structural representations;
Fig. 5 is the embodiment of the invention four data management system structural representations.
Embodiment
The embodiment of the invention provides a kind of data managing method, select the type of memory block according to size of data, with storage in the big file of forming by some memory blocks of the same type, by the record identification of data and the related Search and Orientation of carrying out data of its storage address information.The embodiment of the invention also provides the corresponding data management system of data managing method therewith.Below be elaborated respectively.
Embodiment one, a kind of data managing method, flow process comprise the step of storage as shown in Figure 1, and to the data step of operating of storage, wherein the step of storage comprises:
A1, according to the size of data to be stored, determine the quantity and the type of required memory block, the type of described memory block is divided according to size;
In actual applications, the size distribution characteristics of the data that memory block can be stored as required are divided into several types, and generally speaking, the memory block type is more suitable between 4～16.The big I of memory block of various memory block type correspondences designs than the twice strategy of adjacent less block type according to bigger block type, makes the distribution of block size can cover the distribution of real data size as far as possible uniformly.For example, in the storage of the service end that is used for the communication system operation system to the customer multi-media information data, general big or small characteristics according to multimedia information data in the communication process, can divide 9 memory block types according to 2 power: 4,8,16,32,64,128,256,512,1024 (KB) are according to the needs of practical business, and block type can carry out corresponding modification, increase or deletion.
When the quantity of determining required memory block according to the size of data to be stored and type, can design various concrete storage policies as required, for example, if to reduce disk read-write in the data manipulation (I/O) number of times as far as possible is purpose, then can be when the quantity of determining memory block and type be benchmark with the minimum number, step comprises:
(1), determines the quantity of minimum required memory block according to available memory block type according to the size of data to be stored;
(2) according to the memory block quantity of determining, obtain needed minimum memory block type.
Illustrate: if size of data is 300KB, according to above-mentioned memory block type of giving an example, 1 memory block of minimum only need gets final product, so the required memory block quantity of determining in the step (1) is 1; In memory block quantity is 1 o'clock, and the memory block type of 512KB and 1024KB size all can be finished storage, so the memory block type that definite required minimum memory block type is 512KB in the step (2).
According to above-mentioned storage policy, size of data can be by a memory block storage less than the memory block type of maximum big or small the time, if it is proper that the memory block type is divided, in most of the cases data all can be stored in the memory block, therefore above-mentioned storage policy can reduce the magnetic disc i/o number of times, data write is finished in a magnetic disc i/o as far as possible, under the set prerequisite of magnetic disc i/o performance, improved the data write performance.
Certainly, the design of storage policy also can the disk utilance be attached most importance to, if emphasize to save disk, can comprehensively determine the quantity of memory block to the selection of block type, for example, after determining memory block type, quantity by above-mentioned steps, untapped memory block partly accounts for the memory block size and surpasses certain proportion, then can select adjacent less memory block type.Certainly, take into account magnetic disc i/o efficient, can stipulate that the final memory block number of determining must not surpass the several times of minimum memory block number.Can design multiple storage policy at different purposes in the practical application, select as required by the user.
A2, with data deposit in provide the big file of definite type stores piece, described big file is made up of some memory blocks of the same type;
In the present embodiment, carry out storage with the form of big file.The structure of big file is made up of the memory block of same type as shown in Figure 2, and big file is made up of the memory block of 64KB among Fig. 2, and the side-play amount of all available big filename of each memory block+this memory block in big file represented its unique physical address; In addition, each memory block also can be provided with the integer sign of a correspondence, if one data record needs the polylith storage, this sign can be used to represent the side-play amount of next memory block, like this, for each bar data record, just can represent its unique physical address according to big filename+first block offset.Based on the structure of above-mentioned big file, the step that data is deposited in big file can comprise:
(1) searches the big file of required memory block type correspondence;
(2) has the memory block of distributing respective numbers the big file of enough free block numbers from certain that find;
(3) if all big files of required memory block type correspondence all do not have enough free block numbers, a then newly-built big file that required memory block type is provided distributes the memory block of respective numbers again from described newly-built big file;
(4) the data order is deposited in the distributed store blocks,, also in the integer sign of memory block, preserve the side-play amount of next memory block in described big file if distributed more than one memory block.
For ease of the distribution and the management in space, can stipulate to provide the various big file of dissimilar memory blocks to have consistent size, for example unified is 2G.Certainly, also can adopt miscellaneous stipulations as required, for example stipulate that all big files have consistent memory block quantity etc.
The record identification (ID) of A3, preservation data and the incidence relation of its storage address information;
After storage system deposited a data record in, storage system can be undertaken by Record ID between storage system and the upper-layer service afterwards to its return recording ID alternately in upper-layer service.Business reads, revises, upgrades corresponding data according to Record ID, and storage system is carried out the Search and Orientation of data by Record ID.
Under the simplest situation, whole Record IDs and corresponding storage address information correspondence can be kept in the table, obtain the stored information of data by one-time positioning according to Record ID.
Management for ease of data, a kind of preferred targeting scheme is provided in the present embodiment, that is: the storage address information with data writes concordance list, with concordance list sign add preserve described storage address information be recorded in side-play amount in the described concordance list as Record ID.Can stipulate the record of each index table stores fixed number, identify as concordance list with the concordance list sequence number that the entry of concordance list can be set up new concordance list after using and expiring, the big tables of data of avoiding indefinite length is in managerial difficulty.When searching, determine concordance list by Record ID, determine corresponding record again in this concordance list, can obtain the stored information of data by twice positioning instant, search time, loss was few.
Concordance list can be by Record ID as keyword, and corresponding each Record ID is preserved the storage address information of data, and storage address information can comprise the information of specified data physical address, for example big filename, first block offset, piece number etc.; The information that also can comprise the magnetic disc i/o needs, for example size of data, block size etc.; And other stored informations of data, for example store the date etc.In addition, but the corresponding record ID user mode of preserving this Record ID also in the concordance list for example, represent to delete with 0, this Record ID free time is described, can related new data record; 1 expression effectively illustrates that this Record ID is current related with certain data record, when whole Record ID states of whole concordance lists are 1, if also need to deposit in new data record, just needs to have created new concordance list.
After carrying out storage, the step of data being carried out read-write operation also is provided in the present embodiment, comprising:
The Record ID that A4, resolution data operating sessions provide obtains the storage address information of this Record ID correspondence;
Under the preferred orientation scheme that present embodiment provides, the record position fixing process is:
(1) resolves Record ID, obtain the concordance list and the concordance list side-play amount of this Record ID correspondence;
(2) from the storage address information of the correspondence skew playback record of concordance list.
A5, search corresponding big file and memory block, carry out the data manipulation of being asked according to the storage address information that is write down.
According to storage address information, the if block number is 1, can directly obtain corresponding piece from big file; The if block number can at first obtain first greater than 1, continues to obtain next piece according to first integer sign, up to the data that obtain whole storage.
Embodiment two, a kind of data managing method, the step of its storage is identical with embodiment one, and the difference part is that present embodiment provides the concurrent control to operating sessions when carrying out data manipulation, the flow process of data manipulation comprises as shown in Figure 3 in the present embodiment:
B1, the Record ID that provides according to the data manipulation session are searched corresponding record lock in record lock resource pool, described record lock resource pool comprises several record locks;
Whenever a session of business initiation, when data manipulation was carried out in request, storage system was all locked to record according to the Record ID of business input and is searched corresponding record lock in the resource pool.Based on the storage characteristics of the data record that provides among the embodiment one, can realize the record lock of small grain size very easily.Record in present embodiment lock can simply be packed acquisition by the primary lock that operating system is provided, and primary lock is related with Record ID and corresponding reference count is set, and can carry out control based on Record ID to the session operation requests.The maximum number of the concurrent session that the total amount may command system of locking by controlling recording allows.Resource pool can be locked by the initialization record by system when starting, the Record ID content that all record locks are preserved empties, and reference count puts 0.
B2, judge whether to find the record lock corresponding, if execution in step B3 then, execution in step B4 then if not with searching Record ID;
The reference count of B3, incremental record lock is locked this record according to the data manipulation of being asked and to be added corresponding operable lock, for example read lock or write lock, execution in step B7;
B4, judge whether not the record lock that used by other Record IDs, if execution in step B5 then, execution in step B6 then if not;
B5, getting a record lock that is not used by other Record IDs, write current record ID, increase progressively the reference count of described record lock, is that described record lock adds corresponding operable lock, execution in step B7 according to the data manipulation of being asked;
The data manipulation that B6, hang-up or refusal are asked.
The Record ID that B7, resolution data operating sessions provide obtains the storage address information of this Record ID correspondence;
B8, search corresponding big file and memory block, carry out the data manipulation of being asked according to the storage address information that is write down;
The reference count of B9, the respective record of successively decreasing behind this time conversation end lock when the reference count of record lock is 0, illustrates that this record lock is idle, can distribute to new Record ID.
Owing to may store tens thousand of or hundreds thousand of data record in each big file, the thread-safe when using the concurrent controlling schemes of present embodiment can guarantee to operate simultaneously big file, and improve the concurrent executive capability of system.
System to the embodiment of the invention is elaborated below.
Embodiment three, a kind of data management system 10 as shown in Figure 4, comprise piece memory cell 11, storage policy unit 12, storage performance element 13 and allocation index unit 14, operation execution unit 15;
Piece memory cell 11 is used to provide various dissimilar big files, and every type big file is made up of some memory blocks of the same type, and the type of described memory block is divided according to size;
Piece memory cell in the present embodiment data management system, storage policy unit and storage performance element can be applicable to carry out the step of storage part in embodiment one data managing method, and allocation index unit and operation execution unit can be used for carrying out the step of data operation part.
Embodiment four, a kind of data management system 20 as shown in Figure 5, comprise piece memory cell 21, storage policy unit 22, storage performance element 23 and allocation index unit 24, operation execution unit 25, concurrent control unit 26;
The record identification that provides according to the data manipulation session is provided concurrent control unit 26, searches corresponding record lock in record lock resource pool, and described record lock resource pool comprises several record locks;
If find the record lock corresponding with described record identification, increase progressively the reference count of described record lock, according to the data manipulation of being asked is that described record lock adds corresponding operable lock, allow described operation execution unit 25 to carry out described data manipulation, and the reference count of the described record lock that behind this time conversation end, successively decreases;
If do not find the record lock corresponding with described record identification, then get a record lock that is not used by other record identifications, write the current record sign, increase progressively the reference count of described record lock, according to the data manipulation of being asked is that described record lock adds corresponding operable lock, allow described operation execution unit 25 to carry out described data manipulation, and the reference count of the described record lock that behind this time conversation end, successively decreases;
If do not find the record corresponding lock, and do not have the record lock that is used, then hang up or data manipulation that refusal is asked by other record identifications with described record identification.
The present embodiment data management system can be applicable to carry out the data managing method among the embodiment three.
By the foregoing description as can be seen, the embodiment of the invention adopts the type of selecting memory block according to size of data, with storage in the big file of forming by some memory blocks of the same type, by the record identification of data and the related method of carrying out the Search and Orientation of data of its storage address information; Owing to can select suitable memory block type according to size of data, can try one's best and effectively utilize disk space and improve readwrite performance; Owing to adopt big file to carry out storage, reduced the quantity of documents of in operating system, creating; Because adopt the record identification locate mode related with storage address information, be the constant level positioning time, stable performance can not be subjected to influence that data volume increases and decline significantly.In addition, preferred storage policy, targeting scheme and the concurrency control method etc. that adopt the embodiment of the invention to provide, can better improve the memory property of system, reduce magnetic disc i/o number of times, shortening data search time, the concurrent executive capability of raising system, effectively improve data management efficiency.
More than data managing method that the embodiment of the invention provided and corresponding data management system are described in detail, used specific case herein principle of the present invention and execution mode are set forth, the explanation of above embodiment just is used for helping to understand method of the present invention and core concept thereof; Simultaneously, for one of ordinary skill in the art, according to thought of the present invention, the part that all can change in specific embodiments and applications, in sum, this description should not be construed as limitation of the present invention.
Claims (11)
1, a kind of data managing method comprises the step of storage it is characterized in that the step of described storage comprises:
According to the size of data to be stored, determine the quantity and the type of required memory block, the type of described memory block is divided according to size;
With described data deposit in provide the big file of definite type stores piece, described big file is made up of some memory blocks of the same type;
Preserve the incidence relation of record identification and its storage address information of described data.
2, data managing method according to claim 1 is characterized in that, described size according to data to be stored determines that the quantity of required memory block and the step of type comprise:
According to the size of described data, determine the quantity of minimum required memory block according to available memory block type;
According to the memory block quantity of determining, obtain needed minimum memory block type.
3, data managing method according to claim 1 is characterized in that: described memory block is divided into 4～16 types, and bigger block type is twice than adjacent less block type.
4, data managing method according to claim 1, it is characterized in that, the step of the record identification of described preservation data and the incidence relation of its storage address information is, the storage address information of described data is write concordance list, with concordance list sign add preserve described storage address information be recorded in side-play amount in the described concordance list as described record identification.
5, according to any described data managing method of claim 1～4, it is characterized in that, described with data deposit in provide the step of big file of definite type stores piece comprise:
Search the big file of required memory block type correspondence;
Has the memory block of distributing respective numbers the big file of enough free block numbers from certain that find;
Described data are deposited in the described distributed store blocks in proper order,, also in memory block, preserve the side-play amount of next memory block in described big file if distributed more than one memory block;
The storage address information of described data comprises big filename and the side-play amount of first memory block in described big file.
6, data managing method according to claim 5, it is characterized in that: if all big files of required memory block type correspondence all do not have enough free block numbers, a then newly-built big file that required memory block type is provided distributes the memory block of respective numbers again from described newly-built big file.
According to any described data managing method of claim 1～4, it is characterized in that 7, after the step of described storage, also comprise the step of data manipulation, the step of described data manipulation comprises:
The record identification that the resolution data operating sessions provides obtains the storage address information of this record identification correspondence;
Search corresponding big file and memory block according to described storage address information, carry out the data manipulation of being asked.
8, data managing method according to claim 7 is characterized in that, also comprises before carrying out described data manipulation:
According to described record identification, in record lock resource pool, search corresponding record lock, described record lock resource pool comprises several record locks;
If find the record lock corresponding with described record identification, increase progressively the reference count of described record lock, according to the data manipulation of being asked described record lock is added corresponding operable lock, and the reference count of the described record lock that behind this time conversation end, successively decreases;
If do not find the record lock corresponding with described record identification, then judge whether not the record lock that used by other record identifications, if then get a record lock that is not used by other record identifications, write the current record sign, increase progressively the reference count of described record lock, according to the data manipulation of being asked is that described record lock adds corresponding operable lock, and the reference count of the described record lock that successively decreases behind this time conversation end;
If do not find the record corresponding to lock, and judge and do not have the record lock that is used, the data manipulation of then hanging up or refusing to be asked by other record identifications with described record identification.
9, a kind of data management system is characterized in that, comprising:
The piece memory cell is used to provide various dissimilar big files, and every type big file is made up of some memory blocks of the same type, and the type of described memory block is divided according to size;
The storage policy unit is used for the size according to data to be stored, determines the quantity and the type of required memory block;
The storage performance element, be used for memory block quantity and the type definite according to described storage policy unit, with data deposit in providing of described memory cell the big file of definite type stores piece, and preserve the incidence relation of record identification and its storage address information of described data.
10, data management system according to claim 9 is characterized in that, also comprises:
The allocation index unit is used for the record identification that the resolution data operating sessions provides, and obtains the storage address information of this record identification correspondence according to the incidence relation of described storage performance element preservation;
Operation execution unit is used for searching corresponding big file and memory block according to the storage address information that described allocation index unit obtains, and carries out the data manipulation of being asked.
11, data management system according to claim 10 is characterized in that, also comprises:
Concurrent control unit is used for according to described record identification, searches corresponding record lock in record lock resource pool, and described record lock resource pool comprises several record locks;
If find the record lock corresponding with described record identification, increase progressively the reference count of described record lock, according to the data manipulation of being asked described record lock is added corresponding operable lock, allow described operation execution unit to carry out described data manipulation, and the reference count of the described record lock that behind this time conversation end, successively decreases;
If do not find the record lock corresponding with described record identification, then judge whether not the record lock that used by other record identifications, if then get a record lock that is not used by other record identifications, write the current record sign, increase progressively the reference count of described record lock, according to the data manipulation of being asked is that described record lock adds corresponding operable lock, allows described operation execution unit to carry out described data manipulation, and the reference count of the described record lock that successively decreases behind this time conversation end;
If do not find the record corresponding to lock, and judge and do not have the record lock that is used, the data manipulation of then hanging up or refusing to be asked by other record identifications with described record identification.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200710090676A CN100583832C (en) | 2007-03-30 | 2007-03-30 | Data management method and system |
| PCT/CN2008/070516 WO2008119286A1 (en) | 2007-03-30 | 2008-03-17 | Method and system of data management |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200710090676A CN100583832C (en) | 2007-03-30 | 2007-03-30 | Data management method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101039278A CN101039278A (en) | 2007-09-19 |
| CN100583832C true CN100583832C (en) | 2010-01-20 |
Family
ID=38889919
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200710090676A Active CN100583832C (en) | 2007-03-30 | 2007-03-30 | Data management method and system |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN100583832C (en) |
| WO (1) | WO2008119286A1 (en) |
Families Citing this family (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100583832C (en) * | 2007-03-30 | 2010-01-20 | 华为技术有限公司 | Data management method and system |
| CN101266828B (en) * | 2008-03-25 | 2010-12-01 | 祥硕科技股份有限公司 | Mixed flash memory device and its operation method |
| CN101630992B (en) * | 2008-07-14 | 2013-06-05 | 中兴通讯股份有限公司 | Method for managing shared memory |
| CN101533408B (en) * | 2009-04-21 | 2015-02-04 | 北京四维图新科技股份有限公司 | Processing method and processing device of mass data |
| CN102004700A (en) * | 2010-11-26 | 2011-04-06 | 华为终端有限公司 | Memory space distribution method and device for flash memory |
| CN102609415A (en) * | 2011-01-21 | 2012-07-25 | 腾讯科技（深圳）有限公司 | Method and device for storing read mark and positioning message, and system for positioning message |
| CN102096722B (en) * | 2011-03-21 | 2013-03-27 | 华为数字技术（成都）有限公司 | File storage method and device |
| CN102890677A (en) * | 2011-07-20 | 2013-01-23 | 郑州威科姆科技股份有限公司 | Reliable storage method of non-document system |
| US9047288B2 (en) * | 2012-01-06 | 2015-06-02 | Apple Inc. | Intelligent data delivery and storage based on data characteristics |
| CN103580989B (en) * | 2012-07-31 | 2018-07-24 | 腾讯科技（深圳）有限公司 | Junk mail processing method and system |
| CN104572568B (en) * | 2013-10-15 | 2021-07-23 | 蚂蚁金服（杭州）网络技术有限公司 | Read lock operation method, write lock operation method and system |
| CN105138687B (en) * | 2015-09-17 | 2018-08-10 | 成都索贝数码科技股份有限公司 | It is a kind of that method is described based on the media object for unifying address |
| CN105302912B (en) * | 2015-11-11 | 2018-06-29 | 中国联合网络通信集团有限公司 | A kind of date storage method and equipment |
| CN107967117B (en) * | 2016-10-20 | 2020-10-20 | 杭州海康威视数字技术股份有限公司 | Data storage, reading and cleaning method and device and cloud storage system |
| CN106844421A (en) * | 2016-11-30 | 2017-06-13 | 上海仙剑文化传媒股份有限公司 | A kind of digital picture management method and system |
| CN107346341A (en) * | 2017-07-11 | 2017-11-14 | 王焱华 | A kind of management method of big data |
| CN107395587B (en) * | 2017-07-18 | 2020-01-03 | 北京微星优财网络科技有限公司 | Data management method and system based on multipoint cooperation mechanism |
| CN109164982B (en) * | 2018-08-23 | 2021-04-02 | 京东方科技集团股份有限公司 | Data processing circuit, data processing method and data storage device |
| CN109254733B (en) * | 2018-09-04 | 2021-10-01 | 北京百度网讯科技有限公司 | Method, device and system for storing data |
| CN113835616A (en) * | 2020-06-23 | 2021-12-24 | 华为技术有限公司 | Data management method and system of application and computer equipment |
| CN116846979B (en) * | 2023-08-29 | 2024-03-15 | 江苏睿鸿网络技术股份有限公司 | Scheduling method and scheduling system for resources in cloud computing environment |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6442661B1 (en) * | 2000-02-29 | 2002-08-27 | Quantum Corporation | Self-tuning memory management for computer systems |
| JP2001265628A (en) * | 2000-03-21 | 2001-09-28 | Sanyo Electric Co Ltd | File recording management system |
| US20050268049A1 (en) * | 2002-08-30 | 2005-12-01 | Koninklijke Philips Electronics N.V. | Dynamic memory management |
| JP4176043B2 (en) * | 2004-05-18 | 2008-11-05 | 三洋電機株式会社 | Data recording method and data recording apparatus |
| KR100694069B1 (en) * | 2004-11-29 | 2007-03-12 | 삼성전자주식회사 | Recording apparatus including plurality of data blocks of different sizes, file managing method using the same and printing apparatus including the same |
| CN100583832C (en) * | 2007-03-30 | 2010-01-20 | 华为技术有限公司 | Data management method and system |
-
2007
- 2007-03-30 CN CN200710090676A patent/CN100583832C/en active Active
-
2008
- 2008-03-17 WO PCT/CN2008/070516 patent/WO2008119286A1/en active Application Filing
Also Published As
| Publication number | Publication date |
|---|---|
| WO2008119286A1 (en) | 2008-10-09 |
| CN101039278A (en) | 2007-09-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100583832C (en) | Data management method and system | |
| CN100468402C (en) | Sort data storage and split catalog inquiry method based on catalog tree | |
| CN103064639B (en) | Date storage method and device | |
| CN1295591C (en) | Moving data among storage units | |
| US9047330B2 (en) | Index compression in databases | |
| US9760652B2 (en) | Hierarchical storage architecture using node ID ranges | |
| EP0437159A2 (en) | Method for identifying documents having a particular attribute using a vector relational characteristical object | |
| US20040143597A1 (en) | Digital library system with customizable workflow | |
| CN103020174A (en) | Similarity analysis method, device and system | |
| CN101137981A (en) | Methods and apparatus for managing the storage of content in a file system | |
| CN101315628A (en) | Internal memory database system and method and device for implementing internal memory data base | |
| CN101278289A (en) | System and method for providing an ojbect to support data structures in WORM storage | |
| CN107239569A (en) | A kind of distributed file system subtree storage method and device | |
| CN102739622A (en) | Expandable data storage system | |
| CN107832423A (en) | A kind of file read/write method for distributed file system | |
| CN102609488A (en) | Client, data query method of client, server and data query system | |
| EP3767486A1 (en) | Multi-record index structure for key-value stores | |
| CN102024019A (en) | Suffix tree based catalog organizing method in distributed file system | |
| CN104079600B (en) | File memory method, device, access client and meta data server system | |
| KR20190008228A (en) | A method and system for storing graph information in a database using a B-tree | |
| CN107408239B (en) | Architecture for managing mass data in communication application through multiple mailboxes | |
| CN107506466A (en) | A kind of small documents storage method and system | |
| CN100543748C (en) | A kind of method and system of utilizing file allocation table to carry out the file tracking | |
| US7386692B1 (en) | Method and apparatus for quantized deadline I/O scheduling | |
| CN101261639B (en) | Sharing of database objects |
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 |