CN104111924B - A kind of Database Systems - Google Patents
A kind of Database Systems Download PDFInfo
- Publication number
- CN104111924B CN104111924B CN201310132754.8A CN201310132754A CN104111924B CN 104111924 B CN104111924 B CN 104111924B CN 201310132754 A CN201310132754 A CN 201310132754A CN 104111924 B CN104111924 B CN 104111924B
- Authority
- CN
- China
- Prior art keywords
- hash
- sublist
- subregion key
- unit
- physical library
- 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
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
- G06F16/22—Indexing; Data structures therefor; Storage structures
- G06F16/2228—Indexing structures
- G06F16/2255—Hash tables
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the present invention provides a kind of Database Systems, including：Database server group, comprising multiple database servers, for operation logic database management unit；Logical data base administrative unit, for managing a logical table, logical table includes multiple sublists, each sublist is numbered with sublist, and each sublist is dispersed in different physical library cells；Physical library cells, have storehouse mark, for storing multiple sublists；Data access unit, for calculating target physical library unit, and the target sublist that should be stored in the target physical library unit according to selected access mode and subregion key, the target sublist in access target physical library cells.Increase new physical library and logical table is split, the capacity in single one physical storehouse can reduce after fractionation, and the list that can be accommodated is also more, meet dilatation requirement.
Description
Technical field
The present invention relates to database technology, particularly relates to a kind of Database Systems.
Background technology
The continuous of data increases and is frequently accessed in database, and increasing visit is brought to database server
Ask pressure, this can cause occur bottleneck at CPU, memory or hard disk, and database thus runs machine of slowly even delaying.Solve
Method is typically that dilatation is carried out to database server, and purchase minicomputer, high-end storage hardware, high-end database software and are awarded
Power license (License) quantity.
Some table data of large scale business relational database often increase comparatively fast, and the technology of dilatation mainly includes：Database
Vertical cutting and the horizontal cutting of table.
Data in table are cut into different data files according to certain rule by the horizontal cutting of table, the data text to diminish
Part can improve access speed.If separate unit minicomputer is opened business cutting since business is grown into for bottleneck, using vertical cutting,
Partial service is transferred to new minicomputer and high-end storage, and buys new authorization to run business.Each big data
The cloud database server that storehouse manufacturer releases can also solve scaling concern, but cost is very high.
There are the following problems for the prior art：, it is necessary to buy the minicomputer of corresponding number in dilation process, high-end storage hardware
And authorization, this expense for resulting in the need for is high and can not bear.
The content of the invention
The technical problem to be solved in the present invention is to provide a kind of Database Systems, solution needs to buy in the prior art small-sized
Machine, high-end storage and authorization carry out dilatation, the defects of causing expense high to database.
In order to solve the above technical problems, the embodiment of the present invention provides a kind of Database Systems, including：Database server
Group, comprising multiple database servers, for operation logic database management unit；Logical data base administrative unit, for managing
A logical table is managed, logical table includes multiple sublists, each sublist is numbered with sublist, and each sublist is dispersed in different things
Manage on library unit；Physical library cells, have storehouse mark, for storing multiple sublists；Data access unit, selected for basis
Access mode and subregion key calculate target physical library unit, and target that should be stored in the target physical library unit
Table, the target sublist in access target physical library cells.
In the system, further include：Mode selecting unit, for select HASH modulus mode, section partitioned mode or
Person's uniformity HASH modes access data, and selected access mode is notified the data access unit.
In the system, data access unit includes access processing module, and HASH modulus storage unit and HASH
Modulus reading unit；Access processing module, for obtaining the subregion key included in access request.
In the system, include in HASH modulus storage unit and HASH modulus reading units：The first kind calculates
Module, for when subregion key is integer type, being shifted to subregion key, physical library quantity is used to the subregion key after displacement
Modulus, obtains the storehouse mark in target physical storehouse that should be accessed；Sublist quantity modulus is used to subregion key, obtained remainder is should
When the sublist numbering of the target sublist of access.
In the system, include in HASH modulus storage unit and HASH modulus reading units：Second Type calculates
Module, for when subregion key is character type, carrying out intercity Hash computings to subregion key and obtaining initial hash value, to initial Hash
It is worth shifting function, storehouse mark is obtained using physical library quantity modulus to the hash value after displacement, then initial hash value is used
Sublist quantity modulus obtains sublist numbering.
In the system, include in HASH modulus storage unit and HASH modulus reading units：3rd type calculates
Module, for when subregion key is date type, an integer being converted to using the function of time to subregion key, then by integer type
Mode calculate.
In the system, data access unit further includes：Uniformity HASH storage units and uniformity HASH read single
Member；Include in uniformity HASH storage units and uniformity HASH reading units：Hash module, for being calculated according to subregion key
Go out subregion key cryptographic Hash, and, Hash values are carried out to physics library name and obtain physics library name cryptographic Hash, by physics library name
Cryptographic Hash is distributed on an annulus；Identification module, for subregion key cryptographic Hash to be placed on the annulus, according to subregion key
Find corresponding target physical storehouse in position of the cryptographic Hash on annulus.
In the system, identification module includes：Processing module clockwise, for subregion key cryptographic Hash to be positioned over annulus
Between upper two physics library name cryptographic Hash, the physics library name cryptographic Hash nearest apart from subregion key cryptographic Hash is found clockwise,
The corresponding physics library name cryptographic Hash is target physical storehouse.
In the system, further include：Section zoning unit, for according to corresponding between subregion key and physical library cells
Relation calculates target physical storehouse, wherein, the different value of subregion key correspond to different physical library cells.
In the system, further include：Application server group, the data access unit are located at application server group.
The above-mentioned technical proposal of the present invention has the beneficial effect that：Increase new physical library and logical table is torn open
Point, the capacity in single one physical storehouse can reduce after fractionation, and the list that can be accommodated is also more, meet dilatation requirement, and separate unit is general
Logical PC server can also support its normal operation and handle access.
Brief description of the drawings
Fig. 1 shows the schematic diagram of data base access system；
Fig. 2 represents a kind of flow diagram of method to database server expansion.
Embodiment
To make the technical problem to be solved in the present invention, technical solution and advantage clearer, below in conjunction with attached drawing and tool
Body embodiment is described in detail.
Using the Database Systems that the low-cost server dilatation based on PC is as shown in Figure 1, the Database Systems in the present invention
In：
User terminal, can be specifically computer or mobile terminal etc., be responsible for initiating access request, access request is related to
Operation can be inquiry, increase newly, modification, delete etc..
Application server group, is the access point of access request, has multiple application servers to carry out load balancing in a group,
One of application server performs service logic after receiving access request, according to the contextual analysis of access request and certainly
The fixed database server for specifically handling the access request.
Database, the database mentioned in embodiment, is not specifically noted, that is, refers to logical base.
Logical base, refers to a virtual storehouse, and rear end can correspond to N number of physical library cells.It can be included in one logical base
Multiple logical tables.
Logical table, belongs to some logical base, can correspond in N number of physical library cells, in a physical library cells
M tables are split, i.e. a logical table there are in fact N*M sublists.
Subregion key (UserId), for doing the field of subregion setting, that is, realizes that a field of table computing, the word are divided in point storehouse
Section is the field that must have, and can be integer, character type or date type, the subregion should be included in access request
Key.
Database server group, a database server group are a nodes, including multiple database servers, one
Node uses database server group rather than an only database server, is ensured with host-guest architecture to the normal of database
Access, avoid the access fault that single database server is likely to occur.
The size of one database data volume and the size of every list directly affect the correlated performance of database, split
Database is to eliminate the performance bottleneck of single database server, and the purpose for splitting list is that lifting carries out various operations
Performance.
The embodiment of the present invention provides a kind of Database Systems, as shown in Fig. 2, including：
Database server group, comprising multiple database servers, for operation logic database management unit；
Logical data base administrative unit, for managing a logical table, logical table includes multiple sublists, each sublist tool
There is sublist numbering, each sublist is dispersed in different physical library cells；
Physical library cells, have storehouse mark, for storing multiple sublists；
Data access unit, for calculating target physical library unit according to selected access mode and subregion key, and
The target sublist that should be stored in the target physical library unit, the target sublist in access target physical library cells.
Increase new physical library using the technology provided and logical table is split, whole logical table after fractionation
The sublist that can be accommodated is more, meets dilatation requirement, and the data volume stored in single one physical library unit can reduce, and separate unit is common
PC server can also be supported to realize single one physical library unit normal access.
Business increases can cause very much some database server overload operations soon, cause necessary migrating data, this can wave
Expense human cost simultaneously produces dirty data, so Splitting Logic database management unit and the granularity of logical table will have perspective, need
Be pre-designed needs how many Zhang Zi accommodated in how many a physical library cells and each physical library cells after once splitting
Table.
In a preferred embodiment, the quantity for the sublist stored in each physical library cells is identical.
Physical library quantity N is specifically 100, and sublist quantity M is specifically 100, and working practice shows, this avoids one
Occur splitting again because data volume goes up after secondary fractionation.
Database Systems further include：Application server group, data access unit are located at application server group.Application server
There is network connection between group and database server group.
Data access unit, determines specific physical library cells and sublist according to access request, needs to examine in design process
Whether occur after considering and split again, usually support the selection mode such as HASH modulus or uniformity HASH to access specific thing
Manage library unit and sublist.
In a preferred embodiment, mode selecting unit, for selecting HASH modulus mode or uniformity HASH side
Formula accesses data, and selected access mode is notified the data access unit.
In a preferred embodiment, data access unit includes HASH modulus storage unit, HASH modulus reading units
And access processing module；
Access processing module, for obtaining the subregion key included in access request.
The type of the subregion key defined according to Database Systems carries out different computings, obtains the storehouse mark in target physical storehouse
Numbered with the sublist of target sublist.
In a preferred embodiment, include in HASH modulus storage unit and HASH modulus reading units：
First kind computing module, for when subregion key is integer type, being shifted to subregion key, after displacement
Subregion key uses physical library quantity modulus, obtains the storehouse mark in target physical storehouse that should be accessed；Sublist number is used to subregion key
Mould is measured, obtained remainder is the sublist numbering for the target sublist that should be accessed；
Second Type computing module, for when subregion key is character type, intercity Hash to be carried out to subregion key
(cityhash) computing obtains initial hash value, and to initial hash value shifting function, physics is used to the hash value after displacement
Storehouse quantity modulus obtains storehouse mark, then obtains sublist numbering using sublist quantity modulus to initial hash value；
3rd type computing module, for when subregion key is date type, being converted to subregion key using the function of time
One integer, then calculates in the way of integer type.
The sublist of storehouse mark and target sublist further according to target physical storehouse is numbered, and original SQL language is changed
It is dealt into after writing on target physical storehouse and performs deposit operation.
After access processing module obtains subregion key, HASH modulus reading unit carries out subregion key the fortune as write operation
What is obtained is the storehouse mark and sublist in the target physical storehouse that should be accessed, and carries out rewriting to SQL language and be dealt into object
On Li Ku.
Not only HASH modulus storage unit includes first kind computing module, Second Type computing module and the 3rd type
Computing module, and HASH modulus reading units include：First kind computing module, Second Type computing module and three classes
Type computing module.HASH modulus storage unit and HASH modulus reading unit are obtained in access request at access processing module
Comprising subregion key, and according to first kind computing module, Second Type computing module either the 3rd type computing module.
HASH modulus storage unit and HASH modulus reading unit obtain the target physical obtained at access at access processing module
The storehouse mark in storehouse and the sublist numbering of target sublist access.
In an application scenarios, in HASH modulus modes, according to the son of the storehouse in target physical storehouse mark and target sublist
Table is numbered, and performs read operation.To access the relevant data instance of user, database-name UserDB, DatabaseNum are thing
Storehouse quantity is managed, sublist the quantity TableNum, dbPos of single physical library are storehouse marks, and tablePos is sublist numbering, and subregion key is
UserId, subregion key are int types, obtain the SQL of the basic document of user and are：Select*from Users where
UserId=10001;
Processing procedure includes：
Step 1, Database Systems are parsed to obtain two passes to SQL builds word:The logical table of access is Users, and root
The occurrence 10001 of subregion key is obtained according to the subregion key UserId defined；
Step 2, the physical library where SQL is calculated：dbPos=10001>>Len (N) %/N, that is, obtaining target physical storehouse is：
UserDB_dbPos。
Step 3, the target sublist of SQL is calculated：TablePos=100001%/M, that is, it is Users_ to obtain target sublist
tablePos。
Step 4, corresponding SQL is carried out to be rewritten as:select*from Users_tablePos where UserId=
10001。
Step 5, revised SQL is sent to back-end physical storehouse UserDB_dbPos, front end is returned to after obtaining result.
HASH modulus modes allow Database Systems once to be split dilatation.
Once split if having already been through, database server group has currently managed N number of physical library cells, shows database
System at most can be with subregion on N group physical machines, if N is equal to 100, which can at most be distributed in 100 groups
On physical machine, performance bottleneck will not be run into using so many physical machine operation physical library cells under normal conditions.
If having exceeded capacity limitation again in view of mass data, second of fractionation based on HASH modulus modes is compared
Difficulty, so needing to be planned in project initial period, using the fractionation mode group based on section or based on consistent hashing
Database Systems are built, simplify second of fractionation in later stage.
If have selected a partitioning technique based on section in a database,：
In a preferred embodiment, it is necessary to state that subregion key UserId and subregion key are closed to the corresponding of physical library cells DB
System：
(1-10000]:DB1#UserId 1 to the data less than 10000 on DB1,
(10000-20000]:DB2#UserId 10000 to the data less than 20000 on DB2, round bracket represent bag
Containing (>=or<=), bracket represent do not include (>or<).
Partitioning algorithm based on section, it is only necessary to obtain subregion key, mesh is found in correspondence according to subregion bond energy is enough
Physical library is marked, to the enterprising line access operation in corresponding target physical storehouse.
Dilatation is easier to, and dynamically states new section to new database.Dilatation needs to be newly-increased data
Area's dilatation.
In some scenes, performance bottleneck occurs in some originally existing physical node (database server group),
Need to carry out dilatation to this physical node, consider to build Database Systems using based on consistent hashing subregion.
In a preferred embodiment, data access unit further includes：Uniformity HASH storage units and uniformity HASH
Reading unit；Include in uniformity HASH storage units and uniformity HASH reading units：
Hash module, for calculating subregion key cryptographic Hash according to subregion key, and, Hash is carried out to physics library name and is taken
Physics library name cryptographic Hash is worth to, physics library name cryptographic Hash is distributed on an annulus；
Identification module, for subregion key cryptographic Hash to be placed on the annulus, according to subregion key cryptographic Hash on annulus
Position find corresponding target physical storehouse.
It is specifically to calculate subregion key cryptographic Hash using CityHash algorithms that Hash computings are carried out to physics library name
HashCode。
It can have up to 65535 dummy nodes on annulus, a dummy node correspond to a physics library name Hash
Value.There are correspondence, a physical node between dummy node and physical node (database server group) may correspond to
Multiple dummy nodes a, i.e. physical node can manage multiple physical library cells, for example, a physical node correspond to void
Intend node 1-10000, another physical node correspond to dummy node 10001-20000.It is so designed that to be because of dummy node
Quantity (65535) be physical node finally can dilatation the limit, in other words, the quantity maximum of physical node can be 65535.
These dummy nodes are corresponded to physical node, a physical node correspond to a dummy node section, so calculate
Dummy node fall on some dummy node section, such as when between 10001-20000, it is easy to find corresponding thing
Manage library unit.
Partitioned mode dilatation based on uniformity Hash is supported to original physical node progress dilatation, i.e., some original thing
Reason node receives the data service of this dummy node of 1-10000, and after the physical node dilatation, which only needs to bear
The access of dummy node 1-5000, in the data service of dummy node 5001-10000 originally to newly added physical node,
Computing is simple, and operating lag is minimum.
In an application scenarios, in uniformity HASH modes, there are 100 physical library cells, including：
Step 1, the database-name of 100 physical library cells CMABDB_01 to CMABDB_99 is subjected to Hash value notes
For Hash (cmabdb), physics library name cryptographic Hash is obtained.
Step 2, all physics library name cryptographic Hash are evenly distributed on an annulus, the number range of annulus be [0,
65535]。
Step 3, subregion key cryptographic Hash is calculated according to subregion key.
Step 4, subregion key cryptographic Hash is placed on annulus, subregion key cryptographic Hash finally falls breathes out in two physics library names
Between uncommon value, nearest physics library name cryptographic Hash is found clockwise, which is that access request is accessed
Specific physical library cells.
Uniformity HASH is the data for influencing a database around newly-increased database, after once splitting, still
Can not meet demand when resulting in the need for newly-increased physical library cells, the physics library name cryptographic Hash of newly-increased physical library cells also occurs
On annulus, and two nodes adjacent with oneself on annulus can be only influenced, without influencing all nodes, therefore further expanded
The data migrated during appearance can be relatively smaller.
Dilatation is carried out by the way of split database, the capacity of single one physical library unit can reduce, the common PC clothes of separate unit
Business device can also support its normal operation and handle access.
In an application scenarios, the invention thought that is embodied based on each embodiment is had by taking Directory service as an example
Body illustrates that design principle includes：
1, Directory service needs to provide the storage of address list and various operations, existing 5,000,000,000 communication notes of log
Record, estimates maximum and rises to 30,000,000,000.
2, in the business maximum stage, the number of data lines of sublist should be less than 1,000 ten thousand, i.e. single table it is expected that maximum number of lines is 1
Ten million bar；The mono- tables of the log number of one logical table=physical library quantity * sublist quantity * it is expected maximum number of lines.
3, under normal circumstances, physical library quantity minimum can be 1, and maximum can be 10000, and sublist quantity minimum can be
1, maximum can be 10000, should be from 1,10,100,100,10000 inner selection, and meet physical library quantity * sublist quantity<=
10000。
4, the log number of a logical table is 30,000,000,000, and single table it is expected that maximum number of lines is 1,000 ten thousand, then needs 3000
Zhang Zibiao.
5, it is contemplated that the business that Database Systems will be likely to occur in future increases, and selects 10000 sublists,
6, it is contemplated that be that physical library cells are split to physical machine, the project later stage is possible to dilatation occur, so selection
It is divided into 100 physical library cells, a physical library cells include 100 sublists,
7, Directory service provide with the relevant access of user, so subregion key select UserId.
Idiographic flow includes：
Step 1, address list logical data base includes 100 physical library cells CMABDB_0 ..., CMABDB_99, often
A physical library cells structure is identical, and capacity will be different as business increases, and different user ID includes (as subregion key)
The log of varying number.
Split into 100 physical library cells rather than 2 or 4, be because：
If being split as lesser amt physical library cells, as business growth can cause the load of single physical library unit too
Weight, and the physical library cells position calculated according to User ID may change, and need migration and variation in such cases
Log, raised manpower and resources costs and often brought dirty data, so the quantity of physical library cells should not be very little.
Physical library quantity can increase management service cost too much, according to practical experience be set as 100 it is more suitable.
When log is less, multiple physical library cells are placed on single database server, will as business increases
The larger physical library cells of pressure are accessed to be split on new database server.
Step 2, in physical library cells CMABDB_0 ..., CMABDB_99, address list logical table is split
AddressBook,
By taking physical library cells CMABDB_1 as an example, the identical sublist of newly-built structure in physical library cells CMABDB_1
AddressBook_0 ... ..., AddressBook_99, each user are only possible in a sublist.
Some user communication records are more, and some user communication records are few, and the sublist that this can be caused is larger, and some sublists are small by one
A bit, the various operations to small sublist can more faster, according to design experiences, and sublist quantity maximum is 100.
According to design experiences, when the quantity of log is less than 300W bars in every sublist, inquire about, modification, increases newly and delete
The operating characteristics removed is preferable.
Step 3, a subregion key field-User ID (UserId) is increased in address list logical table AddressBook, this
Field is character string type.Since some heterogeneous systems designed previously have used character string as User ID, so using word
String type is accorded with, to realize the fusion of different isomerization system major key, recommends to use shaping or length in Database Systems afterwards
The User ID of shaping.
Step 4, the User ID in access request, selects the mode orientation and communication such as HASH modulus or uniformity HASH
Physical library cells and sublist where recording.
Step 5, when business, which increases, to be resulted in the need for splitting database, usually there are two fractionation dimensions, first, tearing open
Divide physical library cells, second, tearing sublist open.
Physical library cells are split, horizontal dilatation is realized, physical library cells is migrated, change connection after migration
String configuration.
After splitting sublist, the rule for being routed to sublist in a physical library cells is changed, thus changes ratio
It is more, and tear sublist open in a physical library cells and can not achieve horizontal dilatation.Thus in selection HASH modulus or consistent
In property HASH modes, it can remain that the sublist quantity in single one physical library unit is constant, by increasing physical library cells
Quantity realizes dilatation.
, can be according to record strip number for the business of communication record, field type, the performance of database server determines
Specific appropriate fractionation strategy.
It is using the advantage after this programme：Need not be using the physical library cells and database server of high cost with regard to energy
Realize the storage and access to mass data, split out enough physical library cells in advance, avoid because business growth is too fast
And constantly split again, database volume is reduced after fractionation so that the common PC server of separate unit can also support physical library
The normal operation of unit simultaneously handles access request.
The above is the preferred embodiment of the present invention, it is noted that for those skilled in the art
For, without departing from the principles of the present invention, some improvements and modifications can also be made, these improvements and modifications
It should be regarded as protection scope of the present invention.
Claims (9)
- A kind of 1. Database Systems, it is characterised in that including：Database server group, comprising multiple database servers, for operation logic database management unit；Logical data base administrative unit, for managing a logical table, logical table includes multiple sublists, each sublist has son Table is numbered, and each sublist is dispersed in different physical library cells；Wherein, the logical table belongs to a logical base, corresponds to N In a physical library cells, M tables are split again in a physical library cells, then a logical table there are in fact N*M sublists；Physical library cells, have storehouse mark, for storing multiple sublists；Data access unit, for calculating target physical library unit according to selected access mode and subregion key, and at this The target sublist that should be stored in target physical library unit, the target sublist in access target physical library cells；Data access unit includes access processing module, and HASH modulus storage unit and HASH modulus reading units；Include in HASH modulus storage unit and HASH modulus reading units：Second Type computing module, for when subregion key is character type, carrying out intercity Hash computings to subregion key and obtaining initially Hash value, to initial hash value shifting function, storehouse mark is obtained to the hash value after displacement using physical library quantity modulus, then Sublist numbering is obtained using sublist quantity modulus to initial hash value.
- 2. system according to claim 1, it is characterised in that further include：Mode selecting unit, for selecting HASH modulus mode, section partitioned mode or uniformity HASH modes to access data, And selected access mode is notified into the data access unit.
- 3. system according to claim 2, it is characterised in thatAccess processing module, for obtaining the subregion key included in access request.
- 4. system according to claim 3, it is characterised in that in HASH modulus storage unit and HASH modulus reading units Include：First kind computing module, for when subregion key is integer type, being shifted to subregion key, to the subregion after displacement Key uses physical library quantity modulus, obtains the storehouse mark in target physical storehouse that should be accessed；Subregion key is taken using sublist quantity Mould, obtained remainder are the sublist numberings for the target sublist that should be accessed.
- 5. system according to claim 3, it is characterised in that in HASH modulus storage unit and HASH modulus reading units Include：3rd type computing module, for when subregion key is date type, one to be converted to using the function of time to subregion key Integer, then calculates in the way of integer type.
- 6. system according to claim 3, it is characterised in that data access unit further includes：Uniformity HASH storages are single Member and uniformity HASH reading units；Include in uniformity HASH storage units and uniformity HASH reading units：Hash module, for calculating subregion key cryptographic Hash according to subregion key, and, Hash values are carried out to physics library name and are obtained To physics library name cryptographic Hash, physics library name cryptographic Hash is distributed on an annulus；Identification module, for subregion key cryptographic Hash to be placed on the annulus, according to position of the subregion key cryptographic Hash on annulus Put and find corresponding target physical storehouse.
- 7. Database Systems according to claim 6, it is characterised in that identification module includes：Processing module clockwise, it is suitable for subregion key cryptographic Hash to be positioned on annulus between two physics library name cryptographic Hash Hour hands find the physics library name cryptographic Hash nearest apart from subregion key cryptographic Hash, and the corresponding physics library name cryptographic Hash is target Physical library.
- 8. Database Systems according to claim 1, it is characterised in that further include：Section zoning unit, for calculating target physical storehouse according to the correspondence between subregion key and physical library cells, its In, the different value of subregion key correspond to different physical library cells.
- 9. Database Systems according to claim 1, it is characterised in that further include：Application server group, the data access unit are located at application server group.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310132754.8A CN104111924B (en) | 2013-04-16 | 2013-04-16 | A kind of Database Systems |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310132754.8A CN104111924B (en) | 2013-04-16 | 2013-04-16 | A kind of Database Systems |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104111924A CN104111924A (en) | 2014-10-22 |
| CN104111924B true CN104111924B (en) | 2018-05-11 |
Family
ID=51708719
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310132754.8A Active CN104111924B (en) | 2013-04-16 | 2013-04-16 | A kind of Database Systems |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104111924B (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105426487A (en) * | 2015-11-20 | 2016-03-23 | 北京京东尚科信息技术有限公司 | Distributed database access control method and equipment, distributed database system and expansion method of distributed database system |
| CN105787124A (en) * | 2016-03-28 | 2016-07-20 | 中国建设银行股份有限公司 | Data processing method and device applied to databases as well as data storage system |
| CN105893466B (en) * | 2016-03-28 | 2019-09-20 | 中国建设银行股份有限公司 | A kind of data processing method and system applied to database |
| CN106909597B (en) * | 2016-06-30 | 2020-07-03 | 阿里巴巴集团控股有限公司 | Database migration method and device |
| CN106775470B (en) * | 2016-12-08 | 2020-04-10 | 武汉斗鱼网络科技有限公司 | Data storage method and system |
| CN109213825A (en) * | 2017-06-29 | 2019-01-15 | 北京微影时代科技有限公司 | A kind of data-base capacity-enlarging method and device |
| CN109697815A (en) * | 2019-01-24 | 2019-04-30 | 广州市天河区保安服务公司 | Anti-theft communication network alarming method, appliance arrangement and storage medium |
| CN110059080B (en) * | 2019-04-29 | 2021-12-14 | 珠海天燕科技有限公司 | Data processing method and device |
| CN110955664A (en) * | 2019-12-03 | 2020-04-03 | 中国建设银行股份有限公司 | Method and device for routing messages in different banks and tables |
| CN112181989B (en) * | 2020-10-12 | 2024-09-03 | 浙江大华技术股份有限公司 | Data processing method and device for distributed database |
| CN113268483B (en) * | 2021-05-24 | 2024-07-19 | 北京金山云网络技术有限公司 | Request processing method and device, electronic equipment and storage medium |
| CN114968088B (en) * | 2022-04-08 | 2023-09-05 | 中移互联网有限公司 | File storage method, file reading method and device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101436192A (en) * | 2007-11-16 | 2009-05-20 | 国际商业机器公司 | Method and apparatus for optimizing inquiry aiming at vertical storage type database |
| CN102033938A (en) * | 2010-12-10 | 2011-04-27 | 天津神舟通用数据技术有限公司 | Secondary mapping-based cluster dynamic expansion method |
| CN102968503A (en) * | 2012-12-10 | 2013-03-13 | 曙光信息产业(北京)有限公司 | Data processing method for database system, and database system |
-
2013
- 2013-04-16 CN CN201310132754.8A patent/CN104111924B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101436192A (en) * | 2007-11-16 | 2009-05-20 | 国际商业机器公司 | Method and apparatus for optimizing inquiry aiming at vertical storage type database |
| CN102033938A (en) * | 2010-12-10 | 2011-04-27 | 天津神舟通用数据技术有限公司 | Secondary mapping-based cluster dynamic expansion method |
| CN102968503A (en) * | 2012-12-10 | 2013-03-13 | 曙光信息产业(北京)有限公司 | Data processing method for database system, and database system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104111924A (en) | 2014-10-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104111924B (en) | A kind of Database Systems | |
| KR102462781B1 (en) | KVS tree database | |
| US10417203B2 (en) | Compacting data history files | |
| CN106021266B (en) | Fast multi-tier index supporting dynamic updates | |
| US9639548B2 (en) | Selecting files for compaction | |
| CN103902632B (en) | The method, apparatus and electronic equipment of file system are built in key assignments storage system | |
| CN104077423B (en) | Consistent hash based structural data storage, inquiry and migration method | |
| US10242050B2 (en) | Database caching in a database system | |
| CN102609446B (en) | Distributed Bloom filter system and application method thereof | |
| US20160350302A1 (en) | Dynamically splitting a range of a node in a distributed hash table | |
| JP2015512551A (en) | A consistent ring namespace that facilitates data storage and organization in network infrastructure | |
| CN104424219B (en) | A kind of management method and device of data file | |
| CN110347651A (en) | Method of data synchronization, device, equipment and storage medium based on cloud storage | |
| CN106815267A (en) | Date storage method and device | |
| CN105045850B (en) | Junk data recovery method in cloud storage log file system | |
| CN106570113B (en) | Mass vector slice data cloud storage method and system | |
| CN104054071A (en) | Method for accessing storage device and storage device | |
| CN106960020A (en) | A kind of method and apparatus for creating concordance list | |
| CN115454994A (en) | Metadata storage method and device based on distributed key value database | |
| CN107153512A (en) | A kind of data migration method and device | |
| CN103530067B (en) | A kind of method and apparatus of data manipulation | |
| CN110502472A (en) | A kind of the cloud storage optimization method and its system of large amount of small documents | |
| JP6006740B2 (en) | Index management device | |
| JP2008065716A (en) | Device, method and program for data management | |
| US7822736B2 (en) | Method and system for managing an index arrangement for a directory |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20190621 Address after: 510000 Tianhe District, Guangzhou City, Guangdong Province, 333 Gaotang Road, self-compiled 1-1 Co-patentee after: CHINA MOBILE COMMUNICATIONS GROUP Co.,Ltd. Patentee after: CHINA MOBILE INTERNET Co.,Ltd. Address before: 510000 Tianhe District, Guangzhou City, Guangdong Province, 333 Gaotang Road, self-compiled 1-1 Patentee before: CHINA MOBILE INTERNET Co.,Ltd. Effective date of registration: 20190621 Address after: 510000 Tianhe District, Guangzhou City, Guangdong Province, 333 Gaotang Road, self-compiled 1-1 Patentee after: CHINA MOBILE INTERNET Co.,Ltd. Address before: 510623 13th Floor, Guangdong Global Building, 11 Zhujiang West Road, Zhujiang New Town, Guangzhou City, Guangdong Province Patentee before: CHINA MOBILE GROUP GUANGDONG Co.,Ltd. |