CN106776613A - Data migration method and device - Google Patents
Data migration method and device Download PDFInfo
- Publication number
- CN106776613A CN106776613A CN201510812185.0A CN201510812185A CN106776613A CN 106776613 A CN106776613 A CN 106776613A CN 201510812185 A CN201510812185 A CN 201510812185A CN 106776613 A CN106776613 A CN 106776613A
- Authority
- CN
- China
- Prior art keywords
- child node
- burst
- storage
- node
- stored
- 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.)
- Pending
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
-
- 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 embodiment of the present invention provides a kind of data migration method and device.The method includes：Host node determines the first most child node of storage burst and the second minimum child node of storage burst according to the number that burst is stored in each child node, and each storage burst at least stores a document；Host node stores the number of burst and the difference of the number of storage burst in the second child node in calculating the first child node, if difference is more than first threshold, makes the first child node that the document stored in storage burst to be reduced is sent into the second child node.The embodiment of the present invention determines the first most child node of storage burst and the second minimum child node of storage burst by host node according to the number that burst is stored in each child node, the partial document stored in first child node is sent to the second child node, the data storage of cluster child nodes is set to reach poised state, avoid the big child node of memory data output occur in cluster, improve the computational efficiency of whole cluster.
Description
Technical field
The present embodiments relate to field of computer technology, more particularly to a kind of data migration method and device.
Background technology
With the development of computer technology, distributed data base divides the data in original centralized database
Dissipate storage to multiple by the child nodes of network connection, with obtain bigger memory capacity and Geng Gao's and
Hair visit capacity.
To the process of child node be specially for document to be stored such as XML file storage by prior art：Foundation
The filename of document to be stored calculates cryptographic Hash, according to the scope belonging to the cryptographic Hash from many height of cluster
A child node is determined in node, document to be stored is stored onto the child node, wherein, per height
Node one cryptographic Hash scope of correspondence.If corresponding cryptographic Hash concentrates on one to multiple documents to be stored respectively
In the range of cryptographic Hash, then the plurality of document to be stored will arrive corresponding with the cryptographic Hash scope by storage simultaneously
In child node, cause the memory data output of the child node larger；If or document to be stored is larger will also lead
Cause the memory data output of the child node larger；Because multiple child nodes constitute a cluster, then same collection
The memory data output of different child nodes is different in group.
One big calculating task is divided into multiple small calculating tasks and is distributed to many by the host node in cluster
Individual child node, host node obtains total result of calculation according to the sub- result that each child node is returned, due to collection
The big child node computational efficiency of memory data output is low in group so that the child node returns to sub- result to host node
Time the latest, cause the computational efficiency of whole cluster to decline.
The content of the invention
The embodiment of the present invention provides a kind of data migration method and device, is imitated with the calculating for improving whole cluster
Rate.
The one side of the embodiment of the present invention is to provide a kind of data migration method, including：
Host node determines the first most son of storage burst according to the number that burst is stored in each child node
Node and the second minimum child node of storage burst, each storage burst at least store a document；
The host node is calculated in storing the number of burst and second child node in first child node
The difference of the number of burst is stored, if the difference is more than first threshold, according to the mathematic interpolation institute
State the number of storage burst to be reduced in the first child node；
The host node sends migration instruction to first child node, so that first child node is by institute
State the document stored in storage burst to be reduced and be sent to second child node.
The other side of the embodiment of the present invention is to provide a kind of data migration device, including：
Acquisition module, it is most for determining storage burst according to the number that burst is stored in each child node
The first child node and storage minimum the second child node of burst, each storage burst at least stores a text
Shelves；
Computing module, number and the described second son section for calculating storage burst in first child node
The difference of the number of burst is stored in point, if the difference is more than first threshold, according to the difference meter
Calculate the number of storage burst to be reduced in first child node；
Sending module, for sending migration instruction to first child node, so that first child node
The document stored in the storage burst to be reduced is sent to second child node.
Data migration method provided in an embodiment of the present invention and device, by host node according to each child node
The number of middle storage burst determines second minimum with storage burst of the first most child node of storage burst
Child node, the second child node is sent to by the partial document stored in the first child node, so that the first son section
The number and the difference of the number of storage burst in the second child node that burst is stored in point are less than or equal to first
Threshold value, makes the data storage of cluster child nodes reach poised state, it is to avoid to occur data storage in cluster
The big child node of amount, improves the computational efficiency of whole cluster.
Brief description of the drawings
Fig. 1 is data migration method flow chart provided in an embodiment of the present invention；
Fig. 2 is the structure chart of data migration device provided in an embodiment of the present invention；
The structure chart of the data migration device that Fig. 3 is provided for another embodiment of the present invention.
Specific embodiment
Fig. 1 is data migration method flow chart provided in an embodiment of the present invention.The embodiment of the present invention is for collection
The big child node computational efficiency of memory data output is low in group so that the child node returns to sub- result to host node
Time the latest, cause the computational efficiency of whole cluster to decline, there is provided data migration method, the method
Comprise the following steps that：
Step S101, host node determine storage burst most according to the number that burst is stored in each child node
Many the first child nodes and the second minimum child node of storage burst, each storage burst at least stores one
Document；
In embodiments of the present invention, a cluster includes a host node and multiple child nodes, per height section
Point can include polytype storage burst, and the number of the storage burst of each type is multiple,
Such as any child node can include multiple container A bursts, multiple container B bursts and
Multiple container C bursts, the embodiment of the present invention is just for a type of storage burst (container
C bursts) it is illustrated.Host node calculates cryptographic Hash according to the filename of document to be stored, according to the Kazakhstan
Scope belonging to uncommon value determines a storage burst, and document to be stored is stored onto the storage burst,
Wherein, storage burst one cryptographic Hash scope of correspondence, cryptographic Hash belongs to be stored in same scope
Document will be stored in same storage burst, i.e., one storage burst at least stores a document, according to
All document storages to be stored are specially to the storage burst of each child node according to the method host node
On container C bursts, due to storage burst one cryptographic Hash scope of correspondence, each is to be stored
Document one cryptographic Hash of correspondence, the then number of the storage burst that different child nodes are opened up is different.Host node according to
The first most child node of storage burst and storage point are determined according to the number that burst is stored in each child node
The second minimum child node of piece.
Step S102, the host node calculate and the number of burst are stored in first child node with described the
The difference of the number of burst is stored in two child nodes, if the difference is more than first threshold, according to described
The number of storage burst to be reduced in the first child node described in mathematic interpolation；
Due to the storage burst in the first child node at most, the storage burst in the second child node is minimum, institute
The difference of the number of burst is stored in the number and second child node of stating storage burst in the first child node
More than first threshold, illustrate that the data storage of the cluster child nodes is unbalance, stored in the first child node
Document is more, and the document stored in the second child node is less, and the partial document stored in the first child node is needed
It is transferred to stored in the second child node.Host node in the first child node described in the mathematic interpolation according to treating
The number of the storage burst of reduction.
Step S103, the host node send migration instruction to first child node, so that described first
The document stored in the storage burst to be reduced is sent to second child node by child node.
Host node goes out the number of storage burst to be reduced in first child node according to the mathematic interpolation
Afterwards, migration instruction is sent to first child node, so that first child node is waited to reduce by described
Storage burst in store document be sent to second child node.
In addition, above-mentioned steps S101-S103 may make up a circulation, i.e. step S103 is performed after terminating and returned
Step S101 is returned to, until storing the number and described second of burst described in the cluster in the first child node
The difference that the number of burst is stored in child node is less than or equal to first threshold, illustrates the cluster child nodes
Data storage reach poised state.
The embodiment of the present invention determines storage by host node according to the number that burst is stored in each child node
The first most child node of burst and the second minimum child node of storage burst, will store in the first child node
Partial document be sent to the second child node so that in the first child node store burst number with second son
The difference that the number of burst is stored in node is less than or equal to first threshold, makes the data of cluster child nodes
Storage reaches poised state, it is to avoid occur the big child node of memory data output in cluster, improves whole collection
The computational efficiency of group.
On the basis of above-described embodiment, number of storage burst to be reduced in first child node etc.
In 1/2nd of the difference.For example, the number that burst is stored in first child node is n, institute
The number for stating storage burst in the second child node is m, n-m>K, or n-m>=k, then by the first child node
In (n-m)/2 storage burst in store document move to second child node.
If the size that burst is stored in described each child node is more than or equal to Second Threshold, the storage
Burst is decomposed into two storage bursts；The host node is true according to the number that burst is stored in each child node
Before making the first most child node of storage burst and the second minimum child node of storage burst, also include：
The host node stores the number of burst according to the sampling period in sampling instant obtains each child node.
As document to be stored is constantly deposited into the storage burst of each child node, the size of burst is stored
Constantly increasing, when the size for storing burst is more than or equal to Second Threshold such as 2048MB, this is deposited
Storage burst is decomposed into two storage bursts, such as a certain corresponding cryptographic Hash of burst that stores is 1-100, i.e.,
To be stored document of the cryptographic Hash in the range of 1-100 is stored in the storage burst, when the storage burst
Size when being more than or equal to 2048MB, add up each cryptographic Hash corresponding document since cryptographic Hash is 1
Size, accumulative, such as cryptographic Hash is stopped when first time document size sum is accumulated to more than 1024MB
For the total size of the document in 1-20 is 1020M, cryptographic Hash is that the total size of the document in 1-21 is
1025MB, then it is the corresponding storages point of 1-21 the storage burst to be decomposed into two storage burst i.e. cryptographic Hash
Piece and cryptographic Hash are the corresponding storage bursts of 22-100.Because the number that burst is stored in each child node exists
It is continually changing, then host node stores burst according to the sampling period in sampling instant obtains each child node
Number.
The host node and described each child node belong to same cluster, if increase in the cluster having new
Child node, then using the new child node as second child node.If in addition, the host node and
New child node is increased in the cluster that described each child node is constituted, the new child node is not stored point
Piece, then using new child node as the second minimum child node of storage burst.Using the side of above-described embodiment
The partial document that method will be stored in the first most child node of burst moves to new child node.
The host node and described each child node belong to same cluster, if deleting any from the cluster
Child node, and child node to be deleted is not second child node, then by the child node to be deleted
All documents of middle storage move to second child node；If the child node to be deleted is described
Two child nodes, then move to the 3rd child node by all documents stored in the child node to be deleted,
3rd child node is to store the minimum son section of burst in the cluster in addition to second child node
Point.After all documents migration stored in child node to be deleted terminates, according to above-described embodiment step S101
Rejudge and store in cluster the first most child node of burst and the second minimum child node of storage burst,
And perform subsequent step S102 and S103.
The quantity that the embodiment of the present invention passes through dynamic adjustment cluster child nodes, will store burst most in cluster
Partial document in the first many child nodes moves to new child node, or will be deposited in child node to be deleted
All documents of storage are moved to and store the minimum child node of burst in cluster in remaining child node, and are made
The data storage of the cluster child nodes after state adjustment reaches poised state, improves and cluster is dynamically adjusted
Flexibility.
Fig. 2 is the structure chart of data migration device provided in an embodiment of the present invention.The embodiment of the present invention is provided
Data migration device can perform data migration method embodiment offer handling process, as shown in Fig. 2
Data migration device 20 includes acquisition module 21, computing module 22 and sending module 23, wherein, obtain
Module 21 is used to determine most first of storage burst according to the number for storing burst in each child node
Child node and the second minimum child node of storage burst, each storage burst at least store a document；Meter
In calculating module 22 and being used to calculate and store the number of burst and second child node in first child node
The difference of the number of burst is stored, if the difference is more than first threshold, according to the mathematic interpolation institute
State the number of storage burst to be reduced in the first child node；Sending module 23 is used to be saved to the described first son
Point sends migration instruction, so that the text that first child node will be stored in the storage burst to be reduced
Shelves are sent to second child node.
The embodiment of the present invention determines storage by host node according to the number that burst is stored in each child node
The first most child node of burst and the second minimum child node of storage burst, will store in the first child node
Partial document be sent to the second child node so that in the first child node store burst number with second son
The difference that the number of burst is stored in node is less than or equal to first threshold, makes the data of cluster child nodes
Storage reaches poised state, it is to avoid occur the big child node of memory data output in cluster, improves whole collection
The computational efficiency of group.
The structure chart of the data migration device that Fig. 3 is provided for another embodiment of the present invention.In above-described embodiment
On the basis of, the number of storage burst to be reduced in first child node is equal to two points of the difference
One of.
If the size that burst is stored in described each child node is more than or equal to Second Threshold, the storage
Burst is decomposed into two storage bursts；Acquisition module 21 is additionally operable to according to storage burst in each child node
Before number determines the first most child node of storage burst and the second minimum child node of storage burst,
The number of burst is stored in sampling instant obtains each child node according to the sampling period.
The host node and described each child node belong to same cluster, if increase in the cluster having new
Child node, then using the new child node as second child node.
The host node and described each child node belong to same cluster；Data migration device 20 also includes moving
Shifting formwork block 24, if transferring module 24 is used to delete any child node from the cluster, and it is to be deleted
Child node is not second child node, then move all documents stored in the child node to be deleted
Move on to second child node；If the child node to be deleted is second child node, will be described
The all documents stored in child node to be deleted move to the 3rd child node, and the 3rd child node is institute
State and store the minimum child node of burst in cluster in addition to second child node.
Data migration device provided in an embodiment of the present invention can be specifically for performing what above-mentioned Fig. 1 was provided
Embodiment of the method, here is omitted for concrete function.
The quantity that the embodiment of the present invention passes through dynamic adjustment cluster child nodes, will store burst most in cluster
Partial document in the first many child nodes moves to new child node, or will be deposited in child node to be deleted
All documents of storage are moved to and store the minimum child node of burst in cluster in remaining child node, and are made
The data storage of the cluster child nodes after state adjustment reaches poised state, improves and cluster is dynamically adjusted
Flexibility.
In sum, the embodiment of the present invention passes through host node according to the number that burst is stored in each child node
The first most child node of storage burst and the second minimum child node of storage burst are determined, by the first son
The partial document stored in node is sent to the second child node, so as to store the individual of burst in the first child node
Number is less than or equal to first threshold with the difference of the number of storage burst in the second child node, makes cluster neutron
The data storage of node reaches poised state, it is to avoid occurs the big child node of memory data output in cluster, carries
The computational efficiency of whole cluster high；By the quantity of dynamic adjustment cluster child nodes, will be deposited in cluster
Partial document in storage most the first child node of burst moves to new child node, or by son to be deleted
The all documents stored in node are moved to and store the minimum son section of burst in cluster in remaining child node
Point, and the data storage of the cluster child nodes after dynamic adjustment is reached poised state, improve to collection
The flexibility of group motion state adjustment.
In several embodiments provided by the present invention, it should be understood that disclosed apparatus and method,
Can realize by another way.For example, device embodiment described above is only schematical,
For example, the division of the unit, only a kind of division of logic function, can have in addition when actually realizing
Dividing mode, such as multiple units or component can combine or be desirably integrated into another system, or
Some features can be ignored, or not perform.It is another, shown or discussed coupling each other or
Direct-coupling or communication connection can be the INDIRECT COUPLING or communication link of device or unit by some interfaces
Connect, can be electrical, mechanical or other forms.
It is described as separating component illustrate unit can be or may not be it is physically separate, make
For the part that unit shows can be or may not be physical location, you can with positioned at a place,
Or can also be distributed on multiple NEs.Can select according to the actual needs part therein or
Person whole units realize the purpose of this embodiment scheme.
In addition, during each functional unit in each embodiment of the invention can be integrated in a processing unit,
Can also be that unit is individually physically present, it is also possible to which two or more units are integrated in a list
In unit.Above-mentioned integrated unit can both be realized in the form of hardware, it would however also be possible to employ hardware adds software
The form of functional unit is realized.
The above-mentioned integrated unit realized in the form of SFU software functional unit, can store in a computer
In read/write memory medium.Above-mentioned SFU software functional unit storage is in a storage medium, including some fingers
Order is used to so that a computer equipment (can be personal computer, server, or network equipment etc.)
Or processor (processor) performs the part steps of each embodiment methods described of the invention.And it is foregoing
Storage medium include：USB flash disk, mobile hard disk, read-only storage (Read-Only Memory, ROM),
Random access memory (Random Access Memory, RAM), magnetic disc or CD etc. are various can be with
The medium of store program codes.
Those skilled in the art can be understood that, for convenience and simplicity of description, only with above-mentioned each
The division of functional module is carried out for example, in practical application, as needed can divide above-mentioned functions
With being completed by different functional module, will the internal structure of device be divided into different functional modules, with
Complete all or part of function described above.The specific work process of the device of foregoing description, can be with
With reference to the corresponding process in preceding method embodiment, will not be repeated here.
Finally it should be noted that：Various embodiments above is merely illustrative of the technical solution of the present invention, rather than right
Its limitation；Although being described in detail to the present invention with reference to foregoing embodiments, this area it is common
Technical staff should be understood：It can still modify to the technical scheme described in foregoing embodiments,
Or equivalent is carried out to which part or all technical characteristic；And these modifications or replacement, and
The scope of the essence disengaging various embodiments of the present invention technical scheme of appropriate technical solution is not made.
Claims (10)
1. a kind of data migration method, it is characterised in that including：
Host node determines the first most son of storage burst according to the number that burst is stored in each child node
Node and the second minimum child node of storage burst, each storage burst at least store a document；
The host node is calculated in storing the number of burst and second child node in first child node
The difference of the number of burst is stored, if the difference is more than first threshold, according to the mathematic interpolation institute
State the number of storage burst to be reduced in the first child node；
The host node sends migration instruction to first child node, so that first child node is by institute
State the document stored in storage burst to be reduced and be sent to second child node.
2. method according to claim 1, it is characterised in that wait to reduce in first child node
Storage burst number be equal to the difference 1/2nd.
3. method according to claim 2, it is characterised in that if being stored in described each child node
When the size of burst is more than or equal to Second Threshold, the storage burst is decomposed into two storage bursts；
The host node determines most the of storage burst according to the number that burst is stored in each child node
Before one child node and the second minimum child node of storage burst, also include：
The host node stores the individual of burst according to the sampling period in sampling instant obtains each child node
Number.
4. method according to claim 3, it is characterised in that the host node and each height
Node belongs to same cluster, if increase in the cluster having new child node, by the new child node
As second child node.
5. method according to claim 3, it is characterised in that the host node and each height
Node belongs to same cluster, if any child node is deleted from the cluster, and child node to be deleted is not
It is second child node, then moves to all documents stored in the child node to be deleted described
Second child node；
If the child node to be deleted is second child node, by the child node to be deleted
All documents of storage move to the 3rd child node, and the 3rd child node is except described the in the cluster
The minimum child node of burst is stored outside two child nodes.
6. a kind of data migration device, it is characterised in that including：
Acquisition module, it is most for determining storage burst according to the number that burst is stored in each child node
The first child node and storage minimum the second child node of burst, each storage burst at least stores a text
Shelves；
Computing module, number and the described second son section for calculating storage burst in first child node
The difference of the number of burst is stored in point, if the difference is more than first threshold, according to the difference meter
Calculate the number of storage burst to be reduced in first child node；
Sending module, for sending migration instruction to first child node, so that first child node
The document stored in the storage burst to be reduced is sent to second child node.
7. data migration device according to claim 6, it is characterised in that first child node
In the number of storage burst to be reduced be equal to 1/2nd of the difference.
8. data migration device according to claim 7, it is characterised in that if each height section
When the size that burst is stored in point is more than or equal to Second Threshold, the storage burst is decomposed into two storages
Burst；
The acquisition module is additionally operable to determine storage burst according to the number for storing burst in each child node
Before the first most child nodes and the second minimum child node of storage burst, according to the sampling period in sampling
Moment obtains the number of storage burst in each child node.
9. data migration device according to claim 8, it is characterised in that the host node and institute
State each child node and belong to same cluster, if increase in the cluster having new child node, will be described new
Child node as second child node.
10. data migration device according to claim 8, it is characterised in that the host node and
Described each child node belongs to same cluster；
The data migration device also includes transferring module, if for deleting any sub- section from the cluster
Point, and child node to be deleted is not second child node, then will be deposited in the child node to be deleted
All documents of storage move to second child node；If the child node to be deleted is second son
Node, then move to the 3rd child node by all documents stored in the child node to be deleted, described
3rd child node is to store the minimum child node of burst in the cluster in addition to second child node.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510812185.0A CN106776613A (en) | 2015-11-20 | 2015-11-20 | Data migration method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510812185.0A CN106776613A (en) | 2015-11-20 | 2015-11-20 | Data migration method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106776613A true CN106776613A (en) | 2017-05-31 |
Family
ID=58886034
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510812185.0A Pending CN106776613A (en) | 2015-11-20 | 2015-11-20 | Data migration method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106776613A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112395469A (en) * | 2020-11-27 | 2021-02-23 | 中国银联股份有限公司 | Biological characteristic storage method, device, equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101442435A (en) * | 2008-12-25 | 2009-05-27 | 华为技术有限公司 | Method and apparatus for managing business data of distributed system and distributed system |
| CN101610287A (en) * | 2009-06-16 | 2009-12-23 | 浙江大学 | A kind of load-balancing method that is applied to distributed mass memory system |
| CN101925123A (en) * | 2009-06-12 | 2010-12-22 | 中国移动通信集团设计院有限公司 | Method, system and related functional entities for realizing load balancing in pool area |
| CN103051686A (en) * | 2012-12-10 | 2013-04-17 | 北京普泽天玑数据技术有限公司 | Method and system for isolating dynamic application of distributed system |
-
2015
- 2015-11-20 CN CN201510812185.0A patent/CN106776613A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101442435A (en) * | 2008-12-25 | 2009-05-27 | 华为技术有限公司 | Method and apparatus for managing business data of distributed system and distributed system |
| CN101925123A (en) * | 2009-06-12 | 2010-12-22 | 中国移动通信集团设计院有限公司 | Method, system and related functional entities for realizing load balancing in pool area |
| CN101610287A (en) * | 2009-06-16 | 2009-12-23 | 浙江大学 | A kind of load-balancing method that is applied to distributed mass memory system |
| CN103051686A (en) * | 2012-12-10 | 2013-04-17 | 北京普泽天玑数据技术有限公司 | Method and system for isolating dynamic application of distributed system |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112395469A (en) * | 2020-11-27 | 2021-02-23 | 中国银联股份有限公司 | Biological characteristic storage method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102968498B (en) | Data processing method and device | |
| CN105446979B (en) | Data digging method and node | |
| CN105630955B (en) | A kind of data acquisition system member management method of high-efficiency dynamic | |
| CN102930062B (en) | The method of the quick horizontal extension of a kind of database | |
| CN103345514B (en) | Streaming data processing method under big data environment | |
| KR20160145482A (en) | Method and apparatus of implementing spiking neural network | |
| Xie et al. | Pandas: robust locality-aware scheduling with stochastic delay optimality | |
| CN109408590B (en) | Method, device and equipment for expanding distributed database and storage medium | |
| US10291707B1 (en) | Systems and methods for balancing storage resources in a distributed database | |
| KR20130020050A (en) | Apparatus and method for managing bucket range of locality sensitivie hash | |
| CN107209714A (en) | The control method of distributed memory system and distributed memory system | |
| CN105320773A (en) | Distributed duplicated data deleting system and method based on Hadoop platform | |
| CN112434000B (en) | Small file merging method, device and equipment based on HDFS | |
| CN104702691B (en) | Distributed load equalizing method and device | |
| EP3640813B1 (en) | Cluster-based random walk method and apparatus | |
| CN107133228A (en) | A kind of method and device of fast resampling | |
| CN103970879B (en) | Method and system for regulating storage positions of data blocks | |
| CN110415160B (en) | GPU (graphics processing Unit) topology partitioning method and device | |
| CN108416054A (en) | Dynamic HDFS copy number calculating methods based on file access temperature | |
| CN113835823A (en) | Resource scheduling method and device, electronic equipment and computer readable storage medium | |
| CN107992358A (en) | A kind of asynchronous IO suitable for the outer figure processing system of core performs method and system | |
| CN111522811A (en) | Database processing method and device, storage medium and terminal | |
| Bienkowski | Migrating and replicating data in networks | |
| CN104281636A (en) | Concurrent distributed processing method for mass report data | |
| CN107544848B (en) | Cluster expansion method, apparatus, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170531 |
|
| WD01 | Invention patent application deemed withdrawn after publication |