CN104462124B - Data storing platform method for organizing and data storing platform based on linear Hash table - Google Patents
Data storing platform method for organizing and data storing platform based on linear Hash table Download PDFInfo
- Publication number
- CN104462124B CN104462124B CN201310430856.8A CN201310430856A CN104462124B CN 104462124 B CN104462124 B CN 104462124B CN 201310430856 A CN201310430856 A CN 201310430856A CN 104462124 B CN104462124 B CN 104462124B
- Authority
- CN
- China
- Prior art keywords
- node
- hash table
- cluster
- address number
- hash
- 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
-
- 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
- G06F16/214—Database migration support
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)
- Software Systems (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of data storing platform method for organizing and data storing platform based on linear Hash table, it is related to field of cloud calculation.The present invention extends the structure and address mapping scheme of Hash table, when there is new node to add cluster, keep original mapping between key assignments and address constant, the simply number of significant digit of extended address and Hash table length, assign new caused address to new node, and gradual migrating data while system operation is kept；Similarly, when there is node to leave cluster, keep original mapping between key assignments and address constant, the simply number of significant digit and Hash table length of contraction address, and gradual migrating data while system operation is kept.This method for organizing, the advantages of high by the real-time of Hash table organization cloud cluster was both remained, and made cluster easy to maintain again and extend.The present invention is applied to all cloud storage platforms, the business platforms such as particularly higher to the requirement of real-time, charging that operates on memory database or flow control.
Description
Technical field
The present invention relates to field of cloud calculation, more particularly to a kind of data storing platform method for organizing based on linear Hash table
And data storing platform.
Background technology
With the arriving in big data epoch, cloud mode is the development trend of data storing platform.The method for organizing of cloud storage
Refer to determine method of which data storage in which node according to key assignments, tree-shaped rules of organization and Hash table rules of organization can be divided into
Two kinds.Wherein, Hash table rules of organization are adapted to unirecord inquiry and modification, and advantage is that simple in construction, access speed is fast, and shortcoming is collection
Easy care, increase and deletion of node are not required for stopping cluster running and reorganize data to group.
Method by Hash table organization cloud storage platform is as follows：
1）For the sake of simplified model, it is believed that data have unique key assignments k, and cloud storage platform has N number of node, numbering
For 0~N-1, equity between node.
2）Cloud storage platform maintenance one opens Hash table, and length N, each row of table point to a node, and safeguard one
Hash function H, H (k) result span are between 0~N-1.Hash is a kind of ripe technology in itself, existing Hash
Function can accomplish, the distribution regardless of k, and H (k) distribution is substantially uniform.
3）When there is new data（Key assignments k1）When being loaded into cloud storage platform, cloud storage platform calculates H (k1), it is assumed that and H (k1)=
N, then n nodes are found on Hash table, and the data are stored in n nodes；
4）When user to cloud storage platform request access key value be k1 data when, cloud storage platform also according to H (k1)=
N searches Hash table, and this request is transmitted into n nodes.
Fig. 1 is existing Hash table organisation schematic diagram.As shown in figure 1, if new node adds cluster, it is then
System needs to extend Hash table, its length is reached N+1, the address number of new node is N；Change hash function, new Hash letter
Number H ' (k) span is 0~N；According to new hash function migrating data, for any k, H ' (k) is representated by k
The memory node of data.Data migration process is related to the almost all of node of whole cluster, has to stop cluster clothes for this
Business.
The flow that node leaves cluster is similar to above-mentioned addition cluster, it is also desirable to changes hash function, data migration process
Also relate to the almost all of node of whole cluster.
Have that above-mentioned analysis is visible, easy care, increase and deletion of node are not required for stopping existing Hash table rules of organization cluster
Only cluster runs and reorganizes data, is not suitable for the higher data storing platform of requirement of real-time.
The content of the invention
A technical problem to be solved of the embodiment of the present invention is：Solves cluster existing for existing Hash table rules of organization
Not easy care, increase and deletion of node are required for stopping the problem of cluster runs and reorganizes data.
One side according to embodiments of the present invention, propose a kind of data storing platform organizer based on linear Hash table
Method, including：For the cluster that number of nodes is N, if new node adds the cluster, Hash table length is expanded into N+
1, the address number of new node is designated as N；The partial data on node that Hash table splitting pointer is pointed to moves to address number
For on N new node；One is moved after Hash table splitting pointer, points to next node；Update the value of Hash table digit counterWherein, d represents the value of Hash table digit counter,Expression rounds up.
Another aspect according to embodiments of the present invention, propose a kind of data storing platform tissue based on linear Hash table
Method, including：For the cluster that number of nodes is N+1, if the node that the address number of Hash table least significant end is N leaves
The cluster, Hash table splitting pointer forward one, point to a upper node；Update the value of Hash table digit counterWherein, d represents the value of Hash table digit counter,Expression rounds up；By the section that address number is N
The node that all Data Migrations on point point to Hash table splitting pointer；It is N by Hash table length reduction.
Foregoing data storing platform method for organizing also includes：When user asks the data that access key value is k, according to H (k)
=n calculates k cryptographic Hash, and H represents hash function, and n represents the cryptographic Hash calculated；Obtain the value d of Hash table digit counter；
Interception n rear d positions obtain n '；Interception n rear d-1 positions obtain n "；Compare the position of Hash table splitting pointer and address number n ",
If address number be n " node after Hash table splitting pointer, by request navigate to address number be n " node, such as
Fruit address number be n " node before Hash table splitting pointer, by request navigate to address number be n ' node.
Wherein, the span of hash function is more than the scope of the address number of cluster interior joint.
Another aspect according to embodiments of the present invention, a kind of data storing platform is proposed, including：It is single that node adds processing
Member or/and node leave processing unit；Node adds processing unit for the cluster that number of nodes is N, if new section
Point adds the cluster, Hash table length is expanded into N+1, the address number of new node is designated as N；Hash table splitting pointer is pointed to
Node on partial data move to address number be N new node on；One is moved after Hash table splitting pointer, is pointed to next
Individual node；Update the value of Hash table digit counterWherein, d represents the value of Hash table digit counter,Expression rounds up；Node leaves processing unit for the cluster that number of nodes is N+1, if Hash table least significant end
Address number be N node leave the cluster, Hash table splitting pointer forward one, point to a upper node；Update Hash
The value of epitope counterWherein, d represents the value of Hash table digit counter,Expression rounds up；
The node that all Data Migrations on node for being N by address number point to Hash table splitting pointer；Hash table length is contracted
Small is N.
Data storing platform also includes：Data access processing unit, when asking the data that access key value is k for user,
K cryptographic Hash is calculated according to H (k)=n, H represents hash function, and n represents the cryptographic Hash calculated；Hash table digit is obtained to count
The value d of device；Interception n rear d positions obtain n '；Interception n rear d-1 positions obtain n "；Compare Hash table splitting pointer and address number
N " position, if address number be n " node after Hash table splitting pointer, it is n " that request is navigated into address number
Node, if address number be n " node before the Hash table splitting pointer, it is n ' that request is navigated into address number
Node.
Wherein, data storing platform for example can be cloud storage platform, or other requirement of real-time it is higher, operate in
The business platforms such as charging or flow control on memory database.
The present invention extends the structure and address mapping scheme of Hash table, when there is new node to add cluster, keeps key
Original mapping is constant between value and address, simply the number of significant digit of extended address and Hash table length, and new caused address is assigned
Give new node, and gradual migrating data while system operation is kept；Similarly, when there is node to leave cluster, key assignments is kept
Original mapping is constant between address, simply shrinks the number of significant digit and Hash table length of address, and keeping system operation
Gradual migrating data simultaneously.This method for organizing, the advantages of high by the real-time of Hash table organization cloud cluster was both remained, again
Make cluster easy to maintain and extend.The present invention is applied to all cloud storage platforms, particularly higher to requirement of real-time, operate in
The business platforms such as charging or flow control on memory database.
By referring to the drawings to the present invention exemplary embodiment detailed description, further feature of the invention and its
Advantage will be made apparent from.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments of invention, for those of ordinary skill in the art, without having to pay creative labor, may be used also
To obtain other accompanying drawings according to these accompanying drawings.
Fig. 1 is existing Hash table organisation schematic diagram.
Fig. 2 be the present invention have node add cluster when data storing platform method for organizing schematic flow sheet.
Fig. 3 is the Hash table organisation schematic diagram of the present invention.
Fig. 4 is the data access flow schematic diagram of the present invention.
Fig. 5 is data storing platform organizational process schematic diagram when the present invention has node addition cluster.
Fig. 6 is the schematic flow sheet of the data storing platform method for organizing when present invention has the node to leave cluster.
Fig. 7 is the structural representation of data storing platform one embodiment of the present invention.
Fig. 8 is the structural representation of data storing platform further embodiment of the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.Below
Description only actually at least one exemplary embodiment is illustrative, is never used as to the present invention and its application or makes
Any restrictions.Based on the embodiment in the present invention, those of ordinary skill in the art are not making creative work premise
Lower obtained every other embodiment, belongs to the scope of protection of the invention.
Unless specifically stated otherwise, the part and positioned opposite, the digital table of step otherwise illustrated in these embodiments
Do not limited the scope of the invention up to formula and numerical value.
In order to solve the not easy care of cluster existing for existing Hash table rules of organization, increase and deletion of node are required for stopping
The problem of cluster runs and reorganizes data, the present invention propose a kind of data storing platform organizer based on linear Hash table
Case, in increase and hash function need not be changed during deletion of node, can while system operation is kept gradual migrating data, collect
Group is easy to maintain and extends.The following detailed description of the solution of the present invention.
For the sake of simplified model, it is believed that data have unique key assignments k, and data storing platform has N number of node, address
Numbering is 0~N-1.Data storing platform safeguards a Hash table, and length N, each row of table point to a node, and safeguard
One hash function H, H (k) result span are between 0~N-1.Hash is a kind of ripe technology in itself, existing
Hash function can accomplish, the distribution regardless of k, and H (k) distribution is substantially uniform.Separately below introduce have node add and
Leave data storing platform method for organizing during cluster.
Fig. 2 be the present invention have node add cluster when data storing platform method for organizing schematic flow sheet.
As shown in Fig. 2 data storing platform method for organizing of the present embodiment when there is node to add cluster includes following step
Suddenly：
S201, for the cluster that number of nodes is N, each address of node numbering is 0~N-1, if new section
Point adds the cluster, Hash table length is expanded into N+1, the address number of new node is designated as N.
S202, increase point to Hash table splitting pointer for the Hash table splitting pointer of record data migration situation
Partial data on node is moved on the new node that address number is N.
Fig. 3 shows the Hash table organisation schematic diagram of the present invention.As shown in figure 3, the present invention only need to migrate a node
On data, can be by approximately half of Data Migration to new node.
S203, one is moved after Hash table splitting pointer, point to next node.
S204, increase update Hash table digit counter for the Hash table digit counter of record clustering total size
ValueWherein, d represents the value of Hash table digit counter,Expression rounds up.
Based on the data storing platform method for organizing shown in Fig. 2, when user asks the data that access key value is k, such as Fig. 4
Shown, this method is further comprising the steps of：
S401, k cryptographic Hash is calculated according to H (k)=n.
Wherein, H represents hash function, and k represents the key assignments of data to be accessed, and n represents the cryptographic Hash calculated.In the present invention
In, without changing hash function, the span of hash function can be expressed as 0~2m- 1, the span of hash function（Or
Person says m）It is sufficiently large, it can be ensured that no matter how cluster is extended, and node address is numbered all without beyond this scope, i.e. hash function
Span be more than cluster interior joint address number scope.The rear some positions for intercepting hash function export for Hash, cut
Fetch bit number depends on the circumstances, and is specifically described below.
S402, obtain the value d of Hash table digit counter.
S403, the rear d positions for intercepting n obtain n '.
S404, the rear d-1 positions for intercepting n obtain n ".
S405, compare the position of Hash table splitting pointer and address number n "；
S405a, if address number be n " node after Hash table splitting pointer, request is navigated into address number
For n " node；
S405b, if address number be n " node before Hash table splitting pointer, request is navigated into address number
For n ' node.
In order that the solution of the present invention is clearer, Hash table method for organizing when having the node to add cluster is set forth below
And the example of user accesses data.It will be understood by those skilled in the art that in shown here and discussion all examples, appoint
What occurrence should be construed as merely exemplary, not as limitation.Therefore, the other examples of exemplary embodiment can
With with different values.
Fig. 5 is the data storing platform organizational process schematic diagram when present invention has node addition cluster.
Assuming that cluster has 4 nodes, each address of node numbering is followed successively by 00,01,10,11, at the beginning of Hash table splitting pointer
Beginning, the node that address number is 00 was pointed in position, and now the value d of Hash table digit counter is 2.
Cluster is added if any new node, for extension Hash table length to 5, the numbering for recording the new node is 3 binary
100, because Hash table splitting pointer initial position points to the node that address number is 00, therefore, will be counted accordingly on node 00
According to node 100 is moved to, the data on node 00 probably have half to move to node 100, remaining still on 00, Hash table point
Moved after splitting pointer, point to node 01, now the value d of Hash table digit counter is 3.Because hash function exports the length length of side,
The address number of node 00 is changed into 000.
Such as there is new node to add cluster again, then the new node serial number 101, because now Hash table splitting pointer points to section
Point 01, therefore, the corresponding data on node 01 is moved into node 101, the data on node 01 probably have half to move to section
101 are put, is moved after Hash table splitting pointer, points to node 10, now the value d of Hash table digit counter is still 3.Similarly, node
01 address number is changed into 001.
As there is new node to add cluster again, by that analogy, until the division of node 11 finishes, then Hash table splitting pointer returns
Position all has been subjected to wheel division to 000, now all nodes, and all node address numberings are all 3.
When user asks the data that access key value is k, system first asks H (k)=n, then the value d of fetch bit counter, interception
N rear d positions obtain n ', and n ' rear d-1 positions are referred to as n "；Division pointer is seen again, and such as n " then directly counts this after pointer is divided
N " nodes are navigated to according to request, request is then navigated to n ' nodes by such as n ' ' before pointer is divided.
Fig. 6 is the schematic flow sheet of the data storing platform method for organizing when present invention has the node to leave cluster.
As shown in fig. 6, data storing platform method for organizing of the present embodiment when there is node to leave cluster includes following step
Suddenly：
S601, for the cluster that number of nodes is N+1, if the address number of Hash table least significant end is N node
The cluster is left, Hash table splitting pointer forward one, points to a upper node；
S602, update the value of Hash table digit counterWherein, d represents Hash table digit counter
Value,Expression rounds up；
The node that all Data Migrations on node that address number is N by S603 point to Hash table splitting pointer；
S604, it is N by Hash table length reduction.
Based on the data storing platform method for organizing shown in Fig. 6, user asks the processing processing for the data that access key value is k
Embodiment illustrated in fig. 4 is may be referred to, here only simple general introduction.When user asks the data that access key value is k, counted according to H (k)=n
K cryptographic Hash is calculated, H represents hash function, and n represents the cryptographic Hash calculated；Obtain the value d of Hash table digit counter；Intercept n
Rear d positions obtain n '；Interception n rear d-1 positions obtain n "；Compare the position of Hash table splitting pointer and address number n ", if
The node that address number is n " navigates to the node that address number is n " after Hash table splitting pointer, by request, if ground
The node that location numbering is n " navigates to the node that address number is n ' before Hash table splitting pointer, by request.
As can be seen here, data storing platform organization scheme proposed by the present invention, the structure and address for extending Hash table are reflected
Emission mode, when there is new node to add cluster, original mapping is constant between holding key assignments and address, and simply extended address has
Digit and Hash table length are imitated, assigns new caused address to new node, and gradual transport number while system operation is kept
According to；Similarly, when there is node to leave cluster, keep original mapping between key assignments and address constant, simply contraction address is effective
Digit and Hash table length, and gradual migrating data while system operation is kept.This method for organizing, both remained and passed through
The advantages of real-time of Hash table organization cloud cluster is high, makes cluster easy to maintain again and extend.The data storing platform of the present invention
Suitable for all cloud storage platforms, particularly higher to requirement of real-time, the charging or flow control that operate on memory database
The business platforms such as system.
Based on aforementioned data method for organizing, the present invention also proposes a kind of corresponding data storing platform.Fig. 7 is number of the present invention
According to the structural representation of storage platform one embodiment.As shown in fig. 7, the data storing platform of the present embodiment includes：Node adds
Enter processing unit 701 or/and node leaves processing unit 702.
Node adds a cluster of the processing unit 701 for number of nodes for N, if new node adds the cluster,
Hash table length is expanded into N+1, the address number of new node is designated as N；The portion on node that Hash table splitting pointer is pointed to
Divided data is moved on the new node that address number is N；One is moved after Hash table splitting pointer, points to next node；Renewal
The value of Hash table digit counterWherein, d represents the value of Hash table digit counter,Represent upward
Round.
Node leaves processing unit 702 for the cluster that number of nodes is N+1, if the address of Hash table least significant end
The node that numbering is N leaves the cluster, Hash table splitting pointer forward one, points to a upper node；Update Hash table digit
The value of counterWherein, d represents the value of Hash table digit counter,Expression rounds up；By address
Numbering is the node that all Data Migrations on N node point to Hash table splitting pointer；It is N by Hash table length reduction.
Fig. 8 is the structural representation of data storing platform further embodiment of the present invention.As shown in figure 8, the present embodiment
Data storing platform also includes：
Data access processing unit 803, when asking the data that access key value is k for user, calculate k's according to H (k)=n
Cryptographic Hash, H represent hash function, and n represents the cryptographic Hash calculated；Obtain the value d of Hash table digit counter；Intercept n rear d
Position obtains n '；Interception n rear d-1 positions obtain n "；Compare the position of Hash table splitting pointer and address number n ", if address is compiled
Number for n " node after Hash table splitting pointer, by request navigate to address number be n " node, if address number
For n " node before Hash table splitting pointer, by request navigate to address number be n ' node.
Wherein, the span of hash function can be expressed as 0~2m- 1, the span of hash function（M in other words）
It is sufficiently large, it can be ensured that no matter how cluster is extended, and node address is numbered all without the value beyond this scope, i.e. hash function
Scope is more than the scope of the address number of cluster interior joint.
Aforementioned data storage platform for example can be cloud storage platform, or other requirement of real-time it is higher, operate in
The business platforms such as charging or flow control on memory database.
Data storing platform proposed by the present invention, the structure and address mapping scheme of Hash table are extended, there is new section
When point adds cluster, keep original mapping between key assignments and address constant, simply the number of significant digit of extended address and Hash table length
Degree, assign new caused address to new node, and gradual migrating data while system operation is kept；Similarly, there is node
When leaving cluster, keep original mapping between key assignments and address constant, simply the number of significant digit and Hash table length of contraction address,
And gradual migrating data while system operation is kept.This method for organizing, both remained by Hash table organization cloud cluster
Real-time it is high the advantages of, make cluster easy to maintain again and extend.The data storing platform of the present invention is applied to all cloud storages
Platform, the business platform such as particularly higher to the requirement of real-time, charging that operates on memory database or flow control.
One of ordinary skill in the art will appreciate that hardware can be passed through by realizing all or part of step of above-described embodiment
To complete, by program the hardware of correlation can also be instructed to complete, described program can be stored in a kind of computer-readable
In storage medium, storage medium mentioned above can be read-only storage, disk or CD etc..
The foregoing is only presently preferred embodiments of the present invention, be not intended to limit the invention, it is all the present invention spirit and
Within principle, any modification, equivalent substitution and improvements made etc., it should be included in the scope of the protection.
Claims (10)
1. a kind of data storing platform method for organizing based on linear Hash table, including：
For the cluster that number of nodes is N, each address of node numbering is 0~N-1 in cluster, if new node adds
The cluster, Hash table length is expanded into N+1, the address number of new node is designated as N；
The partial data on node that Hash table splitting pointer is pointed to is moved on the new node that address number is N；
One is moved after Hash table splitting pointer, points to next node；
Update the value of Hash table digit counterWherein, d represents the value of Hash table digit counter,Table
Show and round up.
2. according to the method for claim 1, it is characterised in that also include：
When user asks the data that access key value is k, k cryptographic Hash is calculated according to H (k)=n, H represents hash function, and n is represented
The cryptographic Hash calculated；
Obtain the value d of Hash table digit counter；
Interception n rear d positions obtain n '；
Interception n rear d-1 positions obtain n "；
Compare the position of Hash table splitting pointer and address number n ", if the node that address number is n " refers in Hash table splitting
After pin, by request navigate to address number be n " node, if address number be n " node in Hash table splitting pointer
Before, request is navigated into the node that address number is n '.
3. according to the method for claim 2, it is characterised in that the span of hash function is more than the ground of cluster interior joint
The scope of location numbering.
4. according to the method for claim 1, it is characterised in that the data storing platform is cloud storage platform.
5. a kind of data storing platform method for organizing based on linear Hash table, including：
For the cluster that number of nodes is N+1, each address of node numbering is 0~N in cluster, if Hash table least significant end
Address number be N node leave the cluster, Hash table splitting pointer forward one, point to a upper node；
Update the value of Hash table digit counterWherein, d represents the value of Hash table digit counter,Table
Show and round up；
The node that all Data Migrations on node for being N by address number point to Hash table splitting pointer；
It is N by Hash table length reduction.
6. according to the method for claim 5, it is characterised in that also include：
When user asks the data that access key value is k, k cryptographic Hash is calculated according to H (k)=n, H represents hash function, and n is represented
The cryptographic Hash calculated；
Obtain the value d of Hash table digit counter；
Interception n rear d positions obtain n '；
Interception n rear d-1 positions obtain n "；
Compare the position of Hash table splitting pointer and address number n ", if the node that address number is n " refers in Hash table splitting
After pin, by request navigate to address number be n " node, if address number be n " node in Hash table splitting pointer
Before, request is navigated into the node that address number is n '.
7. a kind of data storing platform, including：Node adds processing unit or/and node leaves processing unit；
Node adds processing unit for the cluster that number of nodes is N, if new node adds the cluster, by Hash table
Length expands to N+1, and the address number of new node is designated as N；The partial data on node that Hash table splitting pointer is pointed to is moved
Move on on the new node that address number is N；One is moved after Hash table splitting pointer, points to next node；Update Hash epitope
The value of counterWherein, d represents the value of Hash table digit counter,Expression rounds up；
Node leaves processing unit for the cluster that number of nodes is N+1, if the address number of Hash table least significant end is N
Node leave the cluster, Hash table splitting pointer forward one, point to a upper node；Update Hash table digit counter
ValueWherein, d represents the value of Hash table digit counter,Expression rounds up；It is N's by address number
The node that all Data Migrations on node point to Hash table splitting pointer；It is N by Hash table length reduction.
8. data storing platform according to claim 7, it is characterised in that also include：
Data access processing unit, when asking the data that access key value is k for user, k Hash is calculated according to H (k)=n
Value, H represent hash function, and n represents the cryptographic Hash calculated；Obtain the value d of Hash table digit counter；Interception n rear d positions obtain
To n '；Interception n rear d-1 positions obtain n "；Compare the position of Hash table splitting pointer and address number n ", if address number is
N " node navigates to the node that address number is n " after Hash table splitting pointer, by request, if address number is n "
Node before Hash table splitting pointer, by request navigate to address number be n ' node.
9. data storing platform according to claim 8, it is characterised in that the span of hash function is more than in cluster
The scope of address of node numbering.
10. data storing platform according to claim 7, it is characterised in that the data storing platform is that cloud storage is put down
Platform.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310430856.8A CN104462124B (en) | 2013-09-22 | 2013-09-22 | Data storing platform method for organizing and data storing platform based on linear Hash table |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310430856.8A CN104462124B (en) | 2013-09-22 | 2013-09-22 | Data storing platform method for organizing and data storing platform based on linear Hash table |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104462124A CN104462124A (en) | 2015-03-25 |
| CN104462124B true CN104462124B (en) | 2018-04-06 |
Family
ID=52908182
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310430856.8A Active CN104462124B (en) | 2013-09-22 | 2013-09-22 | Data storing platform method for organizing and data storing platform based on linear Hash table |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104462124B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106909557B (en) * | 2015-12-23 | 2020-06-16 | 中国电信股份有限公司 | Memory cluster storage method and device and memory cluster reading method and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101676855A (en) * | 2008-09-11 | 2010-03-24 | 美国日本电气实验室公司 | Scalable secondary storage systems and methods |
| CN102457428A (en) * | 2010-10-27 | 2012-05-16 | 中兴通讯股份有限公司 | Load balancing realization method and device for DHT (distributed Hash table) network |
| CN102521304A (en) * | 2011-11-30 | 2012-06-27 | 北京人大金仓信息技术股份有限公司 | Hash based clustered table storage method |
| CN103150394A (en) * | 2013-03-25 | 2013-06-12 | 中国人民解放军国防科学技术大学 | Distributed file system metadata management method facing to high-performance calculation |
| CN103229151A (en) * | 2012-12-27 | 2013-07-31 | 华为技术有限公司 | Partition extension method and device |
-
2013
- 2013-09-22 CN CN201310430856.8A patent/CN104462124B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101676855A (en) * | 2008-09-11 | 2010-03-24 | 美国日本电气实验室公司 | Scalable secondary storage systems and methods |
| CN102457428A (en) * | 2010-10-27 | 2012-05-16 | 中兴通讯股份有限公司 | Load balancing realization method and device for DHT (distributed Hash table) network |
| CN102521304A (en) * | 2011-11-30 | 2012-06-27 | 北京人大金仓信息技术股份有限公司 | Hash based clustered table storage method |
| CN103229151A (en) * | 2012-12-27 | 2013-07-31 | 华为技术有限公司 | Partition extension method and device |
| CN103150394A (en) * | 2013-03-25 | 2013-06-12 | 中国人民解放军国防科学技术大学 | Distributed file system metadata management method facing to high-performance calculation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104462124A (en) | 2015-03-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103514250B (en) | Method and system for deleting global repeating data and storage device | |
| CN105589812B (en) | Disk fragments method for sorting, device and host | |
| CN102255962B (en) | Distributive storage method, device and system | |
| CN105630955B (en) | A kind of data acquisition system member management method of high-efficiency dynamic | |
| US9628438B2 (en) | Consistent ring namespaces facilitating data storage and organization in network infrastructures | |
| CN102221982B (en) | Method and system for implementing deletion of repeated data on block-level virtual storage equipment | |
| CN105975587B (en) | A kind of high performance memory database index organization and access method | |
| CN102968498A (en) | Method and device for processing data | |
| CN110471894A (en) | A kind of data prefetching method, device, terminal and storage medium | |
| CN105683898A (en) | Set-associative hash table organization for efficient storage and retrieval of data in a storage system | |
| CN103701469B (en) | A kind of compression and storage method of large-scale graph data | |
| CN107209714A (en) | The control method of distributed memory system and distributed memory system | |
| CN104809182A (en) | Method for web crawler URL (uniform resource locator) deduplicating based on DSBF (dynamic splitting Bloom Filter) | |
| CN113568562A (en) | Storage system, memory management method and management node | |
| CN101526965B (en) | Locating method of index nodes of disk file and device thereof | |
| CN105138281B (en) | A kind of sharing method and device of physical disk | |
| CN107526550A (en) | A kind of two benches merging method based on log-structured merging tree | |
| CN105159616A (en) | Disk space management method and device | |
| CN105959419A (en) | Establishment method and system for distributed storage structure based on consensus tree | |
| CN107807932A (en) | A kind of hierarchical data management method and system based on path enumeration | |
| CN106570113A (en) | Cloud storage method and system for mass vector slice data | |
| CN102737068A (en) | Method and equipment for performing cache management on retrieval data | |
| CN109407985A (en) | A kind of method and relevant apparatus of data management | |
| WO2024021488A1 (en) | Metadata storage method and apparatus based on distributed key-value database | |
| US9275091B2 (en) | Database management device and database management method |
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 |